Citizen engagement in a digital age : Twitters' role to civic and political mobilizations

(1)

Citizen Engagement in a Digital Age

Twitters’ Role to Civic and Political Mobilizations

Amber Ebrahim 10202234 Master thesis Political Science: International Relations University of Amsterdam, 23 June 2017 Supervised by dr. S. Tanaka Second reader: G. Schumacher amberebrahim@gmail.com Harvard Style (formatting)

(2)

Abstract

Using empirical data, this thesis examines several theories on civic mobilizations and political engagement on social networking platforms. In specific, the impact of three drivers of political mobilization and participation on social networking platforms (political party platforms, hashtags and social capital) will be assessed. By focusing on three subcases of Kinderpardon in the Netherlands, this thesis can fill the empirical and theoretical gap in the existing literature. By constructing the three drivers of political mobilization, the impact of different factors on social networking platforms can be determined and measured more specifically. The results of this thesis indicate a predominantly positive relationship between the use of social networking platforms and political engagement, with informational, expressive, and relational uses being of particular importance.

Keywords: political engagement, civic mobilizations, social networking platforms, political party platforms, hashtags, media-hypes, social capital.

(3)

1. Introduction

Political movements based on digital coordination seem to gather momentum rapidly; yet, many have proved to be unstable and difficult to sustain (Margetts et al. 2016: 4). The Arab Spring in 2014 and the Brazil uprisings in 2014 took most of the world by surprise and pose a challenge to social scientists interested in political communication. Much of the literature in political science is devoted to the study of democratic societies and revolving around the question of whether the government is responsive to the preferences of the mass public. In order for politicians to be responsive to the publics’ preferences and problems, they need to be paying attention to the policy views and debates the public engage with in order to avoid massive uprisings such as the above-mentioned examples. However, the ecology of the online discussions and how the changing use of social media1 affects citizen

engagement and eventually politics is not yet fully researched and understood. This thesis is dedicated to increase the academic understanding of the relation between online discussions and various forms of political mobilizations with political engagement, and the effects of these online discussions on different cases within the same topic of discussion.

Throughout the year, politicians and individuals send millions of messages, called tweets into the world (Parmelee and Bichard 2012: 1). As social networking platforms move from being regarded as little more than a temporary mania to becoming integrated into the everyday life of millions of people, corporations and organizations, empirical research in the social sciences are starting to form a picture of how this widely popular and distributed form of communication affects democracy and political participation by the public (Gustafsson 2010: 3). As people go about their daily lives, they are invited to make “micro-donations” of their personal time and effort to support political causes (Bennett & Margetts 2013, Margetts 2015). These micro-donations include liking, sharing, retweeting, signing petitions, and so on, extending the ladder of participation at the lower end, offering the public to exercise continuous pressure on their democratically elected politicians (ibid). The narrative of this online engagement and the discussions occurring on social media is, in many cases, led by events or happenings that deviate from the status quo. A specific case would be Kinderpardon, emerging in 2011 and engaging the Dutch public online to take part in the debate on whether or not children from

1_{The terms social media, social networking platforms and new media will be used interchangeable in}

(6)

refugees are allowed to stay in the country when their parents are refused asylum or a residence permit. I review three children (referred to as subcases) with different outcomes in this thesis. This leads to the following central research question:

Within a single topic, why do some cases get more attention than others on social networking platforms?

Two sub questions are raised in order to deepen the analysis and to increase the understanding of the relation between social networking platforms and politicians.

How does the use of social networking platforms by the public lead to increased citizen engagement?

Does the increased citizen engagement on social networking platforms increase the level of interest/engagement by politicians?

The theoretical foundation of the questions raised in this introduction is based on literature that connects political civic movements (based on erupting issues and events) in Western democracies with the emergence of social media (Margetts et al. 2016; Rasmussen et al. 2017; Kruikemeier et al. 2014; Barberá et al. 2014; Gustafsson 2010; Freelon et al. 2016; Vasterman 2004). Several theories are attached to this emergence; firstly, collective frames create a dominant societal definition of a problem, therefore increasing the visibility and the gravity of the issue (Vasterman 2004: 41). Collective frames are in this study regarded as hashtags; one or more words characteristic for a specific topic or a subgroup on social media. This in turn increases the agenda setting functionality of social networking platforms, making it more likely that an issue is identified by politicians and eventually added to the political agenda or resulting in a policy success.

Secondly, political activities have shifted from offline activities such as attending meetings in formal organizations towards online engagement (Gustafsson 2010: 7). This knowledge is increasingly important for the relevance of this study. If connecting with the public through social networking platforms increases, politicians have not only the ability to promote themselves, but also to mobilize voters and to have interactive conversations with the public (Kruikemeier et al. 2014).

(7)

I argue that specific events, pushed by different conditions, lead to an increased amount of attention on social networking platforms in the form of online discussions. If the amount of attention increases among the public, a higher number of actors becomes active on the issue. When Dutch politicians get engaged in the online conversation, it becomes more likely that a motion for a policy change will be submitted at the Second Chamber.

In order to test this argument, I analyze a set of factors that condition the impact of online discussions on politicians. First, public opinion might play a stronger role if a higher number of non-governmental actors are actively participating in the online discussion regarding an issue. Second, different types of online and offline connections between people on social networking platforms play a role in the attainment of policy to different degrees (Rasmussen et al. 2017: 3). Third, the use of strong hashtags makes it easier for people to find discussions that are already taking place on social networking platforms, making it easier to get engaged.

As nowadays most communication goes through social media, this will be the main stage of the research. This argument is inherently related to the number of actors active on a particular issue, but also to the matter of influence the public may have on political decision makers. Policy makers are not very likely to follow the advice of actors who act in isolation, but at the same time single messages on social networking platforms will not get the attention of policy makers (Margetts et al. 2016). On several social networking platforms, like Twitter, the amount of attention can quickly take of when it becomes a media-hype; if the public talks about a topic a lot in a short amount of time, events become trending topics.

In this thesis, I assume that public opinion should play an important role in democratic decision-making, as politicians ultimately depend on the public for winning votes to secure re-election (Mayhew 1974; Rasmussen et al. 2017: 5). Indeed, a range of studies provides empirical support for policy responsiveness to public opinion (Erikson et al. 1993; Lax and Philips 2012; Rasmussen et al. 205; Shapiro 2011). However, strong empirical evidence that public opinion on social media influences decision-making in democracies is still in the making.

All tweets regarding the earlier mentioned case of Kinderpardon are analyzed over a 15-month period (between September 1, 2013 to November 1, 2014). Using the Coosto software, all tweets related to this topic are extracted from Twitter. Through this

(8)

method, I will be able to categorize the content of the tweets and make a distinction between the public and the politicians.

Also, the interactive nature of Twitter as a social networking platform where the online discussion takes place, will be analyzed furthermore to understand to what extent the politicians expressed political agenda is affected by what the mass public discuss publicly on this same platform and vice versa. The main advantage of using Twitter as a source of information about online discussions between individuals is that Members of the Dutch Parliament and the mass public are both present on this social networking platform, sending tweets that follow the same format (containing a maximum of 140 characters), and often re-posting each other’s content through retweets.

I will explore which factors lead to an increase in attention on Twitter; are the public and politicians more responsive to specific cases within the main topic, depending on the social ties in the community (indicating whether it makes a difference if a child has a large online and offline activist community in the wings), if the discussion becomes trending (and see if the use of strong hashtags are contributing) or more information and awareness on political party pages (including personal action frames, online petitions, and the content of political debates and opinions)? The results of this thesis indicate a predominantly positive relationship between the use of social networking platforms and political engagement, with informational, expressive, and relational uses being of particular importance.

This thesis specifically contributes to the existing knowledge in the field of political science by increasing further understanding of the roles of social networking platforms in relation to political and civil movements in Western democracies. In addition to contributing to the academic literature on civil mobilizations, this thesis also tests what factors are contributing to political agenda setting through online discussions. What differences can be perceived between one case and another? Empirical evidence on this matter is scarce, as this has not been widely analyzed because most studies focus on mobilizations in non-democratic societies or not include social networking platforms as a unit of analysis.

Furthermore, the second contribution of this thesis is the empirical evidence of mobilizations on the Internet (and Twitter as a social networking platform) to the different cases of Kinderpardon and the involved children. As the research question implies, not every child got the same treatment by the government. This study

(9)

empirically tests what makes the difference and if there is a relation between online mobilizations and different comparable cases, paving the way for future research.

The aim of this thesis is to review existing empirical research on civic mobilizations and political engagement, and to expand the research with respect to the diversity of social media by providing a more robust conclusion on the variation and transformation of roles and relations in social media and political engagement. Social media content analysis provides a more specific qualitative understanding of this subject. Combined with a narrative review, this research improves the conclusion because the data is accessible and subject to replication. My intention is to provide a comprehensive overview of existing literature and theories, and derive generalizable measurements to review the relationship between social media use and political engagement. I aim to shed more light on the linkages between different patterns of social media use and specific forms of political engagement by citizens, such as the use of political party platforms and the interaction between politicians and citizens on social networking platforms, engagement with hashtags and social capital.

The following chapters logically narrate this thesis and the concepts and theories used in this study. In chapter 2, I discuss the existing literature on contemporary Western democracies, Web 2.0, media-hypes and the mobilizing role of social media. Chapter 3 presents the theoretical framework and hypotheses. In this chapter, a theoretical discussion on the impact of social media on democracies and political responsiveness is conducted. Chapter 4 describes the case selection of Kinderpardon and the associated subcases. Chapter 5 introduces the methodology and measurements of the independent variables. Chapter 6 explores the dataset of tweets sent in the timeframe of 15 months by Dutch politicians and the public on Kinderpardon (and the three subcases under review). This thesis concludes in chapter 7 with a summary of the findings, as well as the discussion and limitations of this study, and a list of practical next steps for future research.

2. Literature review

Because of academic interest in the concept of civic movements and political engagement in democratic societies, a vast body of literature exists on this subject. However, for the analysis of public opinion drivers on social media, not all of the existing literature is relevant. In order to limit the scope of this review, the focus is on

(10)

literature that studies the determinants of social networking platforms and how individuals in relation to political activism use these. This includes literature that deals with public opinion on social media in general. This literature review excludes political responsiveness in societies that are not democratic.

2.1 Contemporary Western democracies

The conduct of the state and its policies, public opinion (domestic and international) and all forms of media intermingle in a network of complex relationships. These relationships are being transformed through globalization and the technological marvels that are inherently related to globalization. Considerable academic research has been conducted on these relationships within the fields of political science, and several theorems have been forwarded.

Two perspectives dominate the academic debate: the pluralist model and the elite model (Robinson 2008: 187). The pluralist model assumes that power is dispersed throughout society, including across the media and the public, therefore no group or set of interests dominates. This means that pluralist accounts maintain that media and publics are independent from political influence and therefore can (and should) act as powerful constraints upon governments (ibid).

Conversely, the elite model assumes that power is concentrated within elite groups, who are able to dominate politics and society. Elite accounts maintain that both media and public opinion are subservient to political elites. From this

perspective, media have a rather less independent form of influence – acting as a mouthpiece for government officials, operating to mobilize publics in support of their policies (ibid).

Liberal democracy has empowered populations to scrutinize policy, making public opinion “both an instrument and a factor in the conduct of foreign policy” (Tatu 1984: 26). Not only foreign policy making is subjective to public opinion; democratic principles demand that the government should respond to public opinion, for which elections and media-orated concerns are tools (Robinson 2008: 139). It is, after all, the public that legitimizes the government within democracies. Public opinion should not be thought of as a rational or singular actor, however, nor of singular consensus, rather it is multi-faceted, as the culmination of interest groups and the masses, while ‘attentive opinion’ varies (Hughes 1997: 187).

(11)

I will reflect on how Twitter is structurally influencing our public atmosphere and, in particular, the political debate with reference to Jürgen Habermas' research on "Strukturwandel der Öffentlichkeit". The German philosopher Habermas analyzed how new spaces for public debates sparked in the late eighteenth century in France and the early nineteenth century in England. Newspapers and pamphlets became a prominent way of communicating and formulating public opinion in the seventeenth century. From the late eighteenth century, these media and the meetings of citizens in societies created a new public atmosphere, the so-called "bürgerliche Öffentlichkeit" (Habermas 1990). In the new digital age, media often makes it possible for a thorough restructuring of ways to spread information and ideas. Blogs, web forums, and social media become new spaces for expressing opinions, conducting discussions and the formulation of critical opinions. These media are able to facilitate new forms of political mobilization (Münker 2008: 54).

In the next paragraph, a deeper understanding of these new forms of political mobilization on social networking platforms in relation to Web 2.0 will be conducted.

2.2 Political participation on social networking platforms

Taking a closer look at recent research results on social networking platforms and political participation, the study by Skoric et al. (2016) suggests that there is a positive relationship between the use of social media and political engagement. This relationship is based on the participant’s personality and how the medium is used: it is often those individuals who use social networking platforms primarily to gain information that are politically engaged. The main motivation to join Facebook communities is to socialize and build up reputation among peers (Lilleker and Koc-Michalska 2016: 24), warning us not to become too optimistic about Facebook as a tool for political participation.

These scholars studies the role of social networking platforms in relation to the development and upkeep of bridging and bonding new networks in the form of social capital. Their expectation is that social capital plays a role in fostering civic participation (Skoric et al. 2016: 1820). They found that escapism, diversion and anonymity, characteristics of the former Web 1.0 era, had negative implications for social capital (Shah et al. 2001; ibid). Contrasting to this conclusion, some of the most popular social networking platforms are related to existing offline networks of social

(12)

ties these days, softening the problems present in online interactions such as impoliteness and free-riding (Kavanaugh et al. 2005; Skoric et al. 2016: 1820). Social networking platforms therefore enable individuals to increase interactions with online and offline networks, expanding their networks and increase their social capital (ibid).

Another study by Ahn (2012), for example, found that teenagers use Facebook for both bonding and bridging purposes with regards to social capital, while MySpace use was, for example, associated with bonding social capital only. When comparing this with a study by the same scholar in 2010, interestingly, Ahn’s experiment with Facebook-like social networking platforms created specifically for young individuals failed to find any association with social capital (2010). Ahn concluded that the lack of significant findings is likely due to low participation in such non-mainstream, low-reach social networking platforms. This study suggests that different social networking platforms also mean different technological affordances and user participation across the years. This influences social capital creation, raising further questions regarding possible ways to achieve effective and sustainable political user engagement on these platforms.

In 2014, a dedicated activist movement – Black Lives Matter (BLM) – ignited an urgent national conversation in the U.S. about police killings of unarmed black citizens (Freelon et al. 2016: 5). Social media make it easier for interests to organize; like most contemporary Western social movements, BLM uses online media extensively. In the next section of this literature review, a deeper understanding of why people reach out to digital tools in order to create political turbulence will be conducted by looking at the emergence of Web 2.0 as an interactive online platform and media-hypes on social networking platforms.

2.2.1 Web 2.0

Over the last decades, the increase in number of people with Internet access has been enormous. 46.1% of the world population has access to the World Wide Web in 2016, representing an estimated 3,424,971,237 people. Internet users are defined as individuals who can access the Internet at home, via any device type and connection (internetlivestats.com). The Netherlands has in 2016 15,915,076 Internet users, which corresponds with approximately 93.7% of the population (ibid). In other words, a large majority of the Dutch population uses the Internet.

(13)

A transformation of the Internet’s functionality itself is the results of he global increase of Internet users. The online environment transformed from a passive environment to a more participatory and active environment. This phase of the Internet is often referred to as Web 2.0. What Web 2.0 entails more concretely, is that people have the ability to share information and collaborate to create to build the environment. The environment can be compared with Wikipedia, inducing people to use collaboration as a common effort of creating content, commenting, organizing, sharing and linking to other works. The online environment is networked, meaning that it includes a collection of technologies, such as blogs, social networking sites, microblogs, mashups and so on (Chun et al. 2010: 2).

When political leaders and the mass public engage on Twitter, they are part of the power and promise that comes along with Web 2.0, which inherently means that websites and social networking platforms enable users to create and share their own content with the world (Parmelee and Bichard 2012: 3).

Up until this point I have used the term ‘social networking platforms’ liberally, but it requires a closer definition. Social networking platforms are online platforms where the users – with little or even without interference of professional editors – create their own content through dialogue and interaction (Leyenaar et al. 2012: 323). This definition is in line with the promise of Web 2.0. Twitter is undeniably a large social media platform; ranking tenth on Alexa’s top 500 sites on the web (Alexa.com). With over 313 million unique active users monthly, Twitter makes a viable ground for analyzing the earlier mentioned hypothesis (Twitter 2017).

As social media becomes more prevailing, social science has increasingly turned its attention towards the Internet and its applications as a promise of a more democratic future (e.g. Rheingold 2000; Becker & Slaton 2000) or as a dynamic machine concentrating ever more power into the hands of the few (Van De Donk et al. 1995; Hindman 2008). The development of these applications, in the context of Web 2.0 and social media, combined with “anecdotal evidence of new forms of rapid networked mobilization” (Benkler 2006; Jenkins 2006; Rheingold 2000) created a new interest in the effects of technology on political participation (Gustafsson 2013: 6).

2.3 Media-hypes

In this research, the explosive effects of social networking platforms are explained by the theory on the phenomenon hypes. The definition of the phenomenon

(14)

media-hypes derived from the movements of news waves in itself. According to Peter Vasterman, the media-hype can be defined as “media generated, wall-to-wall news wave, triggered by one specific event and enlarged by the self-reinforcing process within the news production of the media.” (Vasterman 2004: 515). While Vasterman’s study of media-hypes is mainly orientated towards defining the concept in relation to traditional journalistic media, it is also highly applicable to social media. The definition and key dynamics of the media-hype are highly relevant for a qualitative analysis of news waves on Twitter in relation to online activism.

Intense media-hypes can play an important role in the way in which a condition evolves towards a social problem or a crisis (Vasterman 2004: 41). Vasterman explains that social problems are social constructions, emerging in a complex societal force field (ibid). When a new social problem emerges, a small circle of people starts to get involved and raise awareness in order for a political movement to emerge. Central in this process are claim-making activities to convince a larger audience to share their perception of the problem (ibid).

These claim-making activities are perceived as frames – or discourse or collective assumptions – consisting of analyses and perspectives on the social problem. Collective frames – shared frames that are characteristic for communicative outings of a larger group - often emerge in larger mobilizations (ibid: 42). Collective frames highly depend on different forms of media, as they have an important agenda setting functionality, but the influence of a collective frame will reach even further in the case of a media-hype (ibid). The next paragraph will deepen the understanding of the mobilizing role of social media when looking at political engagement by the public.

2.4 The mobilizing role of social media

New information and communication technologies (ICTs) have been perceived as mechanisms for increasing collaborative communication between governments and the public. Especially over the last decade, as democratic systems have become increasingly decentralized, interdependent and linked by new information technologies (Chadwick 2008; Halpern and Gibbs 2013: 1159).

The rise of the Tea Party is one case study of how Twitter can be used to mobilize political activists and give a stage to grassroots movements. Members of the Tea Party (whose main goal is to reduce federal spending) are part of a movement with

(15)

little centralized authority and are spread across the U.S. (Parmelee & Bichard 2012: 12). Yet, without the organizational structure and resources of a major political party, they have staged numerous large protests and elected candidates to office (ibid). Sarno (2009) found that the members of the Tea Party movement use Twitter as an instrument to share their ideas on how to build up a movement and attract people to their protests (ibid).

“Much of the sharing is now facilitated by the fast-growing messaging site Twitter, where today the keyword “teaparty” was one of the most frequently used terms. Users sent out a flurry of updates about attendance, links to photos on Flickr and Photobucket, and videos on YouTube and other sites.” (ibid).

The above mentioned quote illustrates how social networking platforms are used by mobilizing groups to get attention in order to get their ideas to change politics. In addition to the studies on political participation by the public, other academic studies have looked at different legislatures to see who individual politicians are responsive to. Based on social media analysis, Barberá et al. examined primarily whether legislators served co-partisans, or their district more broadly (2014: 3), serving the broader question of overall political government responsiveness. Rather than individual legislator responsiveness, the question has most recently shifted to discussion of whom it is responsive to. Gilens (2012) shows that governments tend to be more responsive to the policy preferences of the wealthy than to the policy preferences of the poor. Ezrow et al. (2011) found that European parties are more responsive to shifts in the mean voter position than to changes in the preferences of their supporters. There are similarly broad and long-standing literatures on how different issues reach the political agenda.

Without doubt, the diffusion of communication media and an increasing level of education have been important factors in the world-wide revival of the movements for civil rights and democracy during the 1980s and the 1990s. The spread of international networks of mass communication and telecommunication had a big impact on the collapse of the Stalinist regimes in Eastern Europe and on the rise of movements fighting democracy in developing countries (Van Dijk 1999: 81).

(16)

Source Country/Region Methods Used Civic engagement drivers

Rasmussen et al. (2017) Europe Social Media Analysis Politicians are active to get votes

Barberá et al. (2016) U.S. Social Media Analysis Politicians are active to get votes

Freelon et al. (2016) U.S. Social Media Analysis & Interviews

Community- and narrative building

Gustafsson (2010) Sweden Interviews Pull/push factors, sharing political affiliation with own private network.

Skoric et al. (2016) - Meta-analysis Online and offline participation to increase social capital.

Table 1: Overview of reviewed case studies

The discussed literature in this chapter has been used to identify the different forms of social media analysis in relation to research on political mobilizations. While previous studies have conducted research on the interaction between the public and politicians on social networking platforms, no general theories can be deducted from these studies. It is not clear whether the same results can be expected in smaller units of individuals responding on Twitter, but also in smaller geographical units such as the Netherlands. This is because the studies lack in clarifying what drives people to engage to participate in online discussions and mobilizations on specific topics. In the next chapter, I will take a closer look at political party platforms, hashtags and social capital as theories to apply and analyze the behavior of individuals with to deduct generalizable interpretations for future research, filling the gap in the current literature through the empirical application to the case of Kinderpardon.

3. Theory

In this chapter, a theoretical discussion on the conditions of media-hypes on social networking platforms that lead to an increase in the amount of attention will be presented. The insights from this section will be used to construct several hypotheses that I will test in in the analysis in chapter 6. First, I will elaborate on political

(17)

engagement and civic engagement on political platforms. Next, I will discuss media-hypes as a theoretical concept. Finally, I will introduce the concepts of online and offline social capital and collective behavior. Every sub-section ends with one or more hypotheses.

3.1 Awareness and information on political platforms

Political engagement can take on many forms, including contributing money to a political party, voting or working for a campaign (Pew Research Center 2014). But on every measure of political engagement, citizen engagement and relations to one’s own ideology and views on democracy lie at heart (Verba et al. 1995: 1). Previous studies have considered numerous ways in which citizens express political participation, leading to the Civic Voluntarism Model, resting on motivation and capacity as the two main factors driving the participatory process (ibid: 3). Gustafsson makes use of the Civic Voluntarism Model (Verba et al. 1995), to understand the drivers of political participation: “resources (time, money, skills), engagement (interest, knowledge, efficacy), and recruitment” (ibid: 269; Gustafsson 2010: 6). He argues that social media has the ability to theoretically influence all three categories (ibid). “Whereas income, education and time in itself would not be affected by using social media, a wider repertoire of political activities (such as various forms of online participation) is available compared to the old way of attending meetings in formal organizations” (ibid; Joyce 2007). These drivers give a basic idea of how political participation emerges in societies, which will be explored in the next paragraphs.

Besides the abovementioned drivers, literature on contemporary democracies identifies media in general as an important factor in shaping civic engagement in the form of grassroots mobilizations (Agarwal et al. 2013). Grassroots mobilizations are occurring when a group of individuals use collective action to organize political changes at a local, national or even an international level. Grassroots refers to building the mobilization from the bottom-up, and is nowadays occurring often on social media. This is supported by several case studies that are mentioned in the former chapter in relation to movements that have manifested on social media over the past years. In order to improve the understanding of mobilizations and civic engagement in relation to social media, my take on this is that networks with a higher certain political party affiliation contribute more to the discussion than networks with a lower affiliation, although networks with a low affiliation (such as

(18)

communities) still add up to be significant. Besides this argument, there is also an incentive for politicians: if connecting with the public through social media increases, legislators and politicians have not only the ability to promote themselves, but also to mobilize voters and to have interactive conversations (Kruikemeier et al. 2014).

Whereas Verba et al. (1995) identify resources, engagement and recruitment as the main drivers of political participation; other scholars contest the link between these theories on political participation, especially on social networking platforms (Agarwal et al. 2014: 327). The dynamics between values, power and technology, recognizing that technology is according to Agarwal et al. (2014) not value-neutral but shaped by the morals and ideas of the creators of technology, fits with the perception of grassroots mobilizations that is used in this thesis (ibid).

However, there are some scholars who question the relation between technology and social movements. According to Beraldo and Galan–Paez (2013) new media blurs the distinction between private and public domains, creating a dilemma for collective action (320; Bimber et al. 2005, 2013). The affordances related to new digital platforms, when effectively in use, can indeed lead up to radical changes in models of citizen participation and models of mobilization (Earl and Kimport 2011; Beraldo and Galan-Paez 2013: 320, 321). Drawing on the ideas of Mancur Olson, it is argued that the “free-rider problem” is the greatest challenge of contemporary political mobilization. Individuals are only likely to take part in political activities if they are provided with incentives to do so (Anstead 2014: 345). New forms of mobilization and organizational logic are connected with the notion of connective action (Bennett and Segerberg 2012). According to them, “whereas traditional collective action requires organizations to provide individuals with incentives for participation, directives for coordination and frames for identification, digitally enabled connective action emerges on the other hand from self-organizing networks of individuals” (Beraldo and Galan-Paez 2013: 321).

The first step in understanding the concept of connective action, is to take a closer look at the core of connective action: personalized communication and its relation to social media (Bennett and Segerberg 2012: 743). Bennett and Segerberg (2012) argue that the level of self-motivation is important for individuals to engage in political action. Due to the globalization, changes have produced a shift in social and political orientations of younger generations in nations that these authors call

(19)

post-industrial democracies (ibid). New, individualized political orientations result in “engagement with politics as an expression of personal hopes, lifestyles or grievances” (ibid). These orientations are expressed through personal action frames. The authors call personal action frames on social networking platforms central to political mobilization. In personalized action frames, “nominal issues may resemble older movement or party concerns in terms of topics (environment, rights, women’s equality, and trade fairness) but the ideas and mechanisms for organizing action become more personalized than in cases where action is organized on the basis of social group identity, membership, or ideology.” (ibid: 744). Personal action frames on social networking platforms – such as “We are the 99%” (Occupy) and “Put people first” (public health care) – are rhetorically inclusive, so every individual can engage with them. Additionally, personal action frames are mediated through social networking platforms, allowing individuals to take ownership of hashtags, slogans and even related memes, and re-use them to reflect their own personalities and affiliations (Bennett & Segerberg 2012: 36–37). Ultimately, this theoretical statement is important, because Bennett and Segerberg explicitly reject any implication that their argument is teleological, noting that connective action is neither more effective than nor a replacement for collective action (Anstead 2014: 345).

In line with this theory on political and civic engagement, I argue that raising awareness and sharing information on political party platforms increases the civic engagement of individuals mobilized on social networking platforms. This could in turn lead to a grassroots mobilization with a higher amount of individuals participating in the online discussion. This leads to the following hypothesis of this thesis:

H1: “Political party platforms will increase civic engagement on social networking platforms, which in turn will lead to a higher amount of individuals participating in the online discussion.”

The next paragraph elaborates on hashtags and personal action frames in the form of media-hypes, forming the second condition of successful use of social networking platforms by the public in order to increase attention for certain topics on social networking platforms.

(20)

3.2 Hashtags

Grassroots mobilisation sometimes seems to explore at random, as most online petitions attract only a small number of supporters in comparison to the discussion, taking place on social networking platforms (Margetts et al. 2016). I argue that this means that success does not depend on the subject matter—similar topics often fare quite differently—but the characteristics of the individual people on social media. Individuals are, for instance, more likely to act by retweeting or signing a petition because they are sensitive to social information: by seeing that others have already signed they know that their endorsement will be seen too (ibid). As a result, if the audience on Twitter participating in the online discussion includes enough people with the same, active mind-set, political hashtags can quickly take off (see figure 1).

Figure 1: Cumulative percentages of messages with strong hashtags on social networking platforms

A theory that identifies a set of criteria for the identification of these media-hypes characterizes the following mechanisms: “a key event; a consonant news wave; a sudden increase in reports on comparable cases; and a strong rise of thematically related news” (Vasterman 2004: 516). Hashtags are considered an important feature of

(21)

media-hypes on social networking platforms, in particular on Twitter (Conover et al. 2011: 90). They allow users to engage by annotating tweets with a label to specify the topic or intended audience of a communication. #BLM (meaning Black Lives Matter) is one example of a topic that has been hyped through the use of a hashtag. Each hashtag identifies a stream of content and a group of individuals interested participating in the online discussion, with users’ tag choices denoting participation in different information channels (ibid).

Messages that spark strong emotion – fear, humor, sadness – are likely to be forwarded and become influential (Parmelee & Bichard 2012: 76). These variables are affecting the topic to become trending and go viral within a short amount of time. Therefore, these variables are included in table 2 as measurements of successful media-hypes in gaining attention.

In line with the theory on media-hypes, this thesis argues that messages with a strong emotion in the network and messages that contain strong hashtags increase attention on topics on social networking platforms. This could in turn lead to trending topics with a higher amount of attention on Twitter. This leads to the following hypotheses of this thesis:

H2: “Messages that contain strong hashtags increase the amount of individuals discussing topics on social networking platforms, which in turn will lead to trending topics with a higher amount of attention in the online discussion.”

H3: “Messages that contain strong emotion in the network increase the visibility of topics on social networking platforms, which in turn will lead to trending topics with a higher amount of attention in the online discussion.”

The next paragraph elaborates on social capital in networks online and offline, forming the third condition of successful use of social networking platforms by the public in order to obtain more attention for a case and eventually, a reaction from politicians.

3.3 Social capital

The concept of social capital in relation to collective identity and political engagement is considered a fundamental feature of political movements among different theoretical perspectives (Polletta and Jasper 2001). These scholars challenge,

(22)

as illustrated in paragraph 3.1, that political engagement on social networking platforms experience a free-riding issue. In the constructivist school of thought, part of the definition of social movements: is that the construction of a collective identity is the process to account for when studying social movements, it being related to the production of new meanings and challenging dominant perspectives (Melucci 1996). The theory of informational movements explicitly states the crucial role and the power associated with collective identities within a network society (Castells 1997).

Building on the idea of informational movements and network societies, the size of a network undeniably plays an important role as one of its most relevant characteristics. The reach of a network largely determines its power and usability (van Dijk 1999: 80). Accessibility for the interconnected people on the social media platforms also makes one of the strong features of a network – a feature that is largely present on Twitter – but at the same time one of its weaknesses. The chances of ignorant and unauthorized people having access to the network increase proportionally, because the true identity of people behind online profiles remains often unknown, especially on social networking platforms as Twitter, where usernames invite individuals to use a nickname instead of their real names.

Some social media campaigns or personal action frames are perceived as having a lack of an identifiable leadership. But people at the front are still needed – be it the early adopters. It is expected to find that highly influential people are at the front of successful mobilizations, hashtags or petitions. However, Margetts et al. (2016) found, among other researchers, that it is nothing is as influential as huge numbers of people without very many followers. It is them that are playing the crucial role, the mass public (ibid.). It could be that an influential is involved, but it would be surprising if this person is leading the online discussion.

Earlier conducted studies have found causal effects of engagement on social media on political knowledge in empirical investigations. These studies explain the findings with the “surprise effect” of unexpected political social media content, neutralizing the effect of already politically interested people actively searching for political information on the Internet (Gustafsson 2010: 7).

To further understand the mechanism of engagement on social media, other scholars have established a strong connection between political participation and social capital; in particular, the relation between participation and weak ties. Mark S.

(23)

Granovetter (1973) discovered that “people rarely act on mass-media information, unless it is also transmitted through personal ties; otherwise one has no particular reason to think that an advertised product or an organization should be taken seriously.” (1374; Gustafsson 2010: 7). As the number of weak ties increases, so does the likelihood of participation (Teorell 2003). If a person’s direct social network is considered large, the chance that he or she will be asked to engage in a political movement is higher (ibid). “Organizing weak ties in social network sites allows for an individual to stay connected to brief acquaintances also when moving to another geographical area, thereby expanding the network and increasing the possibility of recruitment.” (Gustafsson 2013: 7).

Thus, in line with the theories on large communities and social capital, this study argues that a higher size of the online and offline community will lead to increasing attention for a particular child in the case of Kinderpardon. This leads to the following hypothesis of this study:

H4: “Offline and online communities with strong ties will lead to a higher amount of attention in online discussions.”

Based on the discussed theories in this chapter, it can be assumed that information and awareness through political party platforms, the use of hashtags and emotion in messages and social capital frames of the topic and the matter of political engagement in the network. Eventually, this would have an effect on reactions of politicians, leading to agenda setting or even policy successes as subcategories for political responsiveness. Figure 2 provides an overview in the form of a conceptual model of the theoretical expectations as presented in this chapter. In chapter 6, the hypotheses will be tested using different data sources.

(24)

Figure 2: Conceptual model

4. Case Study

This chapter will outline the case study of Kinderpardon and the organization Defence for Children supporting the mobilizations related to this cause. First, I will underline the importance of this case study and why this case study was picked for this thesis. Next, I will elaborate on the process of Kinderpardon over the past years and the relation with the organization Defence for Children. Finally, I will introduce the subcases that are used in the analysis to measure the determinants of online discussions to see what influences the successes of online discussions and if a variation in attention from politicians can be perceived.

The outcome of online discussions and campaigning on social networking platforms is not simply a function of the characteristics of the actors at play themselves, but also of the characteristics of the policy issues at stake and the preference alignment of other actors. Because the success of the individuals in online mobilizations depends, for instance, on how conflicting the environment in which the discussion takes place is. Mobilizations for a cause should be regarded as a “collective enterprise” in which individuals rely not only on themselves but also on the actions and characteristics of likeminded actors to achieve their (policy) goals. This thesis would underline the importance of taking the potentially crucial role of the public into account when

(25)

studying online mobilizations in democracies not only for normative reasons – because political decision makers should be responsive to citizens – but also to obtain an accurate picture in the power that the public may have over policy and decision-making processes in the contemporary age.

Defence for Children

Defence for Children is an international organization that is fighting for children’s rights in the Netherlands, based on the UN Rights of the Child treaty (Defenceforchildren.nl 2017a). Defence for Children is the Chairman of Defence for Children International, an international network that includes 47 national sections and members. Defence for Children uses lobbying, research, advisement, information provision, education and action to achieve their goals (ibid).

The organization does not accept funding from the government or political parties, but they do collaborate together with the Dutch government and the European Commission to stimulate human beings all over the world in several projects and missions.

The political agenda of “children’s rights” has seen a rapid growth since the rise of numerous organizations concerned with this issue. Political mobilization on social media regarding the issue of Kinderpardon has contributed to this growth, and would form the main case of this thesis. Kinderpardon is heavily related to online discussions on behalf of children from refugees who did not get a Dutch citizenship after applying for a permanent residency. Kinderpardon is meant for children of refugees who have been living in the Netherlands for a long period of time. This settlement has been reached after years of advocacy, which mainly took place on social media.

After years of battling for the rights of rooted children has come to a successful deconstruction on 31 October 2012. The government announced a Children's Pardon for children of asylum seekers and single-minded minors who have been in the Netherlands for more than five years. In addition, there was a definitive arrangement for children who will be rooted in the Netherlands in the future and not have a residence permit. The arrangement for long-term residents has come into force on February 1, 2013. (Defence for Children 2017e)

(26)

While Kinderpardon is an official settlement nowadays, not all children of refugees are allowed to stay. Online discussions still plays a big role. A piece of the puzzle that this thesis is seeking for is the question of why some children can stay, while others have to leave. The mentioned hypotheses in the former chapter will function as a guideline to research what makes a case successful.

Subcases

Doritty (Ghana), Tri (Vietnam) and Wime (Angola) are children who are either born in the Netherlands or who have lived in the Netherlands for most of their lives. Still, they are (initially) excluded from Kinderpardon because their parents applied for another license instead of asylum, because the situation in their home country is considered to be safe or because they have not been under the supervision of the Government for a selected period of time.

Doritty was born and raised in The Hague. Dorrity's mother is from Ghana and came to The Netherlands in 1998 (Defence for Children 2017b). Dorrity went to elementary school during the period of analysis and had many friends. Doritty is very ambitious, later she wants to become a lawyer. Every Sunday, Dorrity goes to church with her mother and sister Anna, and they get a lot of support. The case of Doritty is still under consideration by the government.

The parents of Tri are Vietnamese. In 2001, they fled to the Netherlands and applied for asylum (Defence for Children 2017c. Tri was born in the Netherlands and lived the first years of his life in an asylum seekers center. Tri grew up in Wageningen and was still living in this city in the period of analysis. He was attending group seven of elementary school and performing well. In 2016, Tri and his family received a residence permit to stay in the Netherlands; therefore indicating an exception on the policy was made.

At the age of ten, Wime came to the Netherlands by himself (Defence for Children 2017d). Born in Angola, Wime grew up in the Netherlands. At the time of the analysis, Wime was 22 years old, and waiting for his verdict in jail. An exception was made for Wime on the policy, leading to him staying in the Netherlands.

These subcases will be further analyzed in chapter 6 by comparing online discussions, petitions, hashtags and collective frames, and the results of the efforts of people.

(27)

5. Methodology

In this chapter, a theoretical discussion on the conditions of media-hypes on social media that lead to a reaction from politicians will be presented. The insights from this section will be used to construct several hypotheses and variables that I will test in in the analysis in chapter 6. First, I will elaborate on the data and case study choices that I made. The third paragraph elaborates on the process of data gathering. The fourth paragraph identifies the population in the empirical dataset. The fifth paragraph introduces the dependent and independent variables once more, and concretely state how these will be measured. The final paragraph in this methodology is about ethical considerations.

5.1 Data

In defending this form of social media analysis as academically relevant, it is sensible to justify the data first. It can be argued that academic research in the social sciences related to topics like these, often lean towards a quantitative approach. However, online discussions occurring on social networking platforms, such as Twitter or Facebook, can be extended beyond its “intended” use as data by linking the available information from people’s profiles or the content they actually post to the case study.

For this study, I employ a case study method, which is well suited for ‘how’ and ‘why’ questions of complex social phenomena, which are best understood in real-life contexts (Gerring 2007; Agarwal et al. 2016: 329). This case with its subcases is chosen because Kinderpardon can illustrate the mentioned conditions and give insights on different types of online mobilizations and the outcomes. The subcases are relevant in this mobilization, because they exemplify the border cases of the newly adopted policy, which means cases of children that along the margins of the policy. This means that some children are not eligible for Kinderpardon, because they do not fit the bureaucratic rules that the IND set for Kinderpardon. Given the qualitative character of this study, it has to be taken into account that results are not generalizable into every context imaginable. The results however are of utmost importance for understanding how online discussions are exercising pressure on politicians in democracies, and what kind of reactions can be expected. In the following section, background information on Defence for Children will be provided, with a focus on their communication and publications. Next, the specific contexts of

(28)

Kinderpardon in the period of analysis will be explored. Finally, social media analysis as the main data collection method applied in this thesis will be explained and discussed.

5.2 Research design

For this thesis, I selected three sub-cases in the main case of Kinderpardon. The similar sub-cases are consisting of three children who were initially rejected Kinderpardon during the period of analysis. Doritty (Ghana), Tri (Vietnam) and Wime (Angola) allow me to capture and describe the different processes behind different cases within the same perimeters in greater detail than considering the entire period of analysis as a single unit, and compare the different outcomes (as elaborated in chapter 4) with the observed measurements.

I test if communication on political party platforms is a suitable measurement by checking if there are perceived relationships between the discussion on social networking platforms and reaction by politicians (agenda setting and handing in bills for example). By looking at secondary data on the attitude of politicians towards social media, pages from political parties on Kinderpardon and a semi-structured interview with Linda Voortman, I can control for an interrelationship when comparing these with the dataset and the trends in the different periods of time.

The hashtags of Kinderpardon and the cases analyzed in the online discussion will be dealt with by looking into which Twitter accounts started the discussion in the early beginning of the time frame and the hashtags that are being used. What are the tweets saying?

Social capital in this thesis is analyzed by looking if relationships between petitions, traditional media and the outcomes of the subcases can be perceived. The semi-structured interview with Linda Voortman will be used in this paragraph once more to identify and understand what is perceived.

5.3 Data collection

The analysis in this thesis is based on data that has been collected in two ways. First, a qualitative social media content analysis of Twitter and the web and weblogs of Defence for Children and Kinderpardon is conducted in order to enable insights on

(29)

how individuals and organizations use the Internet to discuss their issues or concerns online. Content analyses are conceptualized as “careful, detailed, systematic examinations and interpretations of particular bodies of material in an effort to identify patterns, themes, biases, and meanings.” (Berg and Lune 2012: 349). While some researchers in the social sciences differentiate between qualitative and quantitative research methods, Berg and Lune categorize content analysis in the mixed methods section, stating that it can be both – qualitative and quantitative – at the same time (ibid: 354).

The first step in formulating the categories of analysis is by consulting the theoretical framework in chapter 3. Table 2 shows the categories based on the theoretical expectations as drivers of online discussions and mobilizations, based on the literature and the relevant theories mentioned in this thesis: civic engagement, media-hypes and collective behavior.

(30)

Political attitudes Media-hypes Social capital (online and offline)

Websites _• _{Updates on}

political agenda

• Motion and bill

request in the Second Chamber • Online petitions • Online informational videos • News publications

Twitter _• Discussion in the form of comments and arguments on the topic • Messages that contain strong emotion • Powerful hashtags • Retweeting and mentioning

• Messages that link

to information on topics, actions and petitions

• Personal action

frames

Table 2: Categories of the social media analysis

Besides the qualitative social media content analysis, this study draws on secondary sources to give more guidance to the collected data. Interpretation of the content can say what is perceived, but not why. Surveys would have been suitable and complementary to this data and analysis; however, the cases go back to 2013. This is why I made the decision to analyze online petitions and their trend lines with the trend lines perceived in the social media analysis. A Dutch research agency (Weber Shandwick) conducted in 2013, the same period of the analysis, a general research on the attitude of Dutch politicians towards social networking platforms in their daily work. These results are also used to give depth to the social media analysis and to provide a better understanding of underlying motivations of politicians to engage with individuals and organizations on Twitter and in online discussions.

Besides the secondary data from Weber Shandwick, I requested three raw data sets from petities.nl, each containing data on when individuals signed petitions for the three children under review in this thesis. Minor data cleaning was necessary in the data sets containing petition data; I manually summed up petition numbers per day (as every signature was separately recorded) and I switched columns in order to make graphical representations of the data.

(31)

For the same reason, I conducted two semi-structured interview, one with the owner of petities.nl and the other one with Linda Voortman, a Dutch politician in the Groenlinks party. The interviews have been conducted in April and June 2017, and were carried out face-to-face.

5.4 Population and sample

I obtained the 43,881 tweets analyzed in the thesis through Coosto from Twitter. These tweets include all public postings between September 1, 2013 and November 1, 2014, matching at least one of the keywords in the query and have not been deleted as of June 14, 2017. Tweets of deleted profiles or retweets from deleted tweets are included in this dataset.

The population at display in this thesis is people living in the Netherlands, Dutch natives, living in local districts. Therefore, only Dutch people or people living in the Netherlands and communicating in Dutch are included in this dataset. By checking the IP address individuals used to send tweets, I tracked their location. The dataset consists of 12,169 unique authors. I applied a spam and location filter to ensure that all accounts in this dataset are in fact Dutch (speaking) citizens located in the Netherlands2.

5.5 Measurement instruments

In order to answer the questions raised in the former chapters and test the hypotheses and to analyze the impact of drivers of the amount of attention on social networking platforms, the three dimensions (political party platforms, hashtags and social capital) need to be operationalized. These variables are selected on the basis of the theoretical framework in Chapter 3.

5.5.1 Dependent variable

Attention for a specific case on social networking platforms: The dependent variable of this thesis is ‘attention on social networking platforms’. As I mentioned before, this study is researching three subcases, so this dependent variable is case specific. I measure the amount of attention by capturing all of the tweets sent in the period of analysis.

(32)

5.5.2 Independent variables

Political party platforms: to measure how political party platforms are contributing to an increasing amount of attention for topics on Twitter, I look at the political party affiliation in the form of memberships of society. By acknowledging a relationship between political parties and their members, I will measure social capital by indicating what political parties say on their pages regarding Kinderpardon and how (or if) they act on the policy, how people respond to this content on Twitter and if correlations between these activities can be indicated in the empirical data.

This is also measured by a semi-structured interview with Linda Voortman (Groenlinks), of the politicians engaging with the conversation on Kinderpardon on Twitter.

Hashtags: The independent variable “strong hashtags” was measured by indicating if a hashtag followed the next principles, based on the theoretical framework: clear, short, transparent, and related to a campaign or an individual.

Social capital: I will measure social capital by looking at the number of online petitions being signed, and if there are correlations between the signatures per day and the online discussions. Besides this, I will check if offline activities (such as charity events) influence the amount of attention on Twitter.

5.6 Ethical considerations

Besides these steps, I realize that some users may object to their username or tweets being used in this thesis. I take full responsibility for any objections in relation to including them in this research. My ethical choices are therefore seeking for a balance between the rights of all individuals participating in the online conversation and the right of the public and scholars to explore how movements rise to prominence on social networking platforms. I treated the information at hand with respect and discretion. However, if anyone whose name or content is referenced to in this thesis is experiencing harm or objections, I encourage them to contact me so I can reduce or eliminate the reference.

Citizen engagement in a digital age : Twitters' role to civic and political mobilizations