Democratic deliberation on social network sites: a study of digital deliberative discourse in the 2012 election

Democratic deliberation has been studied in various diverse environments, however, scholars have yet to examine its characteristics when conducted on political campaigns' social network sites. The present study sought to fill this gap by exploring the contextual issues that shape campaigns' and citizens' deliberative experiences on this interactive digital venue. A phenomenological theoretical approach was used to frame this research and to craft the research design that involved analysis of data collected from two sources: multimedia text published on social network sites and interviews with individuals involved with the operation of political campaigns' social network site. Analysis of these two sources reveals that citizens participate in deliberative discourse using various strategies that are distinct to social network site technology. Specifically deliberators presented hyperlinks, personal identities, ideological beliefs, and facts about candidates' past experiences to support their opinions. Additionally, citizens developed their arguments by drawing on content that was disseminated by political campaigns, other deliberators, and media advertisements. Results also describe characteristics concerning the relationship between political campaigns and the democratic deliberation engaged on their social network sites. It was concluded that campaigns influence the nature of deliberation through the strategic operation of their social network sites. Additionally, while campaigns in the present study did not use citizen deliberation to influence policy making, campaigns widely agreed that this iv discourse was valuable and should be used more broadly to influence the larger political arena. Based on the major findings that emerged, this empirical research argues that political campaign social network sites cultivate valuable deliberative discourse that can be used to inform formal governing procedures and subsequently influence broader democratic processes.

DEMOCRATIC DELIBERATION ON SOCIAL NETWORK SITES: A STUDY OF DIGITAL DELIBERATIVE DISCOURSE IN THE 2012 ELECTION by Stephanie E. Bor A dissertation submitted to the faculty of The University of Utah in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Communication The University of Utah May 2013 Copyright © Stephanie E. Bor 2013 All Rights Reserved STATEMENT OF DISSERTATION APPROVAL The dissertation of STEPHANIE E. BOR has been approved by the following supervisory committee members: ROBERT K. AVERY , Chair March 8, 2013 Date Approved MATTHEW J. BURBANK , Member March 8, 2013 Date Approved TIMOTHY L. LARSON , Member March 8, 2013 Date Approved JOY Y. PIERCE , Member March 8, 2013 Date Approved DAVID J. VERGOBBI , Member March 8, 2013 Date Approved and by KENT A. ONO , Chair of the Department of COMMUNICATION and by Donna M. White, Interim Dean of The Graduate School. ABSTRACT Democratic deliberation has been studied in various diverse environments, however, scholars have yet to examine its characteristics when conducted on political campaigns' social network sites. The present study sought to fill this gap by exploring the contextual issues that shape campaigns' and citizens' deliberative experiences on this interactive digital venue. A phenomenological theoretical approach was used to frame this research and to craft the research design that involved analysis of data collected from two sources: multimedia text published on social network sites and interviews with individuals involved with the operation of political campaigns' social network site. Analysis of these two sources reveals that citizens participate in deliberative discourse using various strategies that are distinct to social network site technology. Specifically deliberators presented hyperlinks, personal identities, ideological beliefs, and facts about candidates' past experiences to support their opinions. Additionally, citizens developed their arguments by drawing on content that was disseminated by political campaigns, other deliberators, and media advertisements. Results also describe characteristics concerning the relationship between political campaigns and the democratic deliberation engaged on their social network sites. It was concluded that campaigns influence the nature of deliberation through the strategic operation of their social network sites. Additionally, while campaigns in the present study did not use citizen deliberation to influence policy making, campaigns widely agreed that this iv discourse was valuable and should be used more broadly to influence the larger political arena. Based on the major findings that emerged, this empirical research argues that political campaign social network sites cultivate valuable deliberative discourse that can be used to inform formal governing procedures and subsequently influence broader democratic processes. As a scholar it is my first and foremost objective to contribute to achieving social justice in society, and I dedicate this work to underrepresented voices that share my passion to create positive change in democratic political systems. TABLE OF CONTENTS ABSTRACT....................................................................................................................... iii LIST OF FIGURES ......................................................................................................... viii ACKNOWLEDGEMENTS............................................................................................... ix Chapters I INTRODUCTION ...................................................................................................1 Background to the Study..........................................................................................2 Genesis and Justification of the Problem.................................................................4 Statement of the Problem.........................................................................................5 Definition of Terms .................................................................................................5 Organization of the Dissertation ..............................................................................8 II LITERATURE REVIEW......................................................................................11 Democratic Deliberation........................................................................................11 Digital Democracy and Deliberation .....................................................................19 Social Network Sites in Political Campaigning.....................................................27 III RESEARCH DESIGN...........................................................................................35 Phenomenology .....................................................................................................35 Data........................................................................................................................38 Procedures..............................................................................................................45 IV RESULTS ..............................................................................................................60 Strategies Used by Citizens to Engage in Democratic Deliberation .....................60 Relationship Between Campaigns and Citizen Deliberation.................................88 V SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS ......................110 Summary..............................................................................................................110 Discussion............................................................................................................115 Limitations ...........................................................................................................125 vii Suggestions for Future Research .........................................................................128 APPENDICES ................................................................................................................132 Consent Form ......................................................................................................132 Interview Guide ...................................................................................................135 REFERENCES ................................................................................................................138 LIST OF FIGURES Figure Page 1 Screen shot of Matheson's campaign Facebook Page...................................................57 2 Screen shot of Love's campaign Facebook Page ..........................................................58 3 Example of Facebook "post" and "comments" .............................................................59 ACKNOWLEDGEMENTS I am grateful to all of the elements that I have encountered in the universe that have inspired me to complete my doctoral dissertation. I am immensely grateful to Dr. Bob Avery, chair of my doctoral committee, for supporting me through every step of my program. Beginning with encouraging me to apply to the Ph.D. program at the University of Utah, Dr. Avery has continued to be the most reliable, inspirational, and kind mentor that any graduate student could ask for. I will treasure his advice and emulate his professional poise and strength throughout my future career. I am also greatly appreciative of my committee members, each of whom imparted distinct pieces of wisdom and support that have helped me complete my dissertation and degree. Thank you to Dr. Matthew Burbank who so willingly accepted and appreciated my desire to embark on interdisciplinary work by introducing me to theories and research from the field of political science. Thank you to Dr. Tim Larson who stimulated my interest in marketing communication and always offered an interesting and novel perspective regarding the topics and people I study. Thank you to Dr. Joy Pierce who I feel lucky to have had the opportunity to work with in both teaching and research. I deeply admire Dr. Pierce's ability to demonstrate a x mix of sophistication, creativity, and compassion in academia and I am grateful for the chance to learn from her. Thank you to Dr. David Vergobbi who consistently challenged me to think beyond the text and to recognize relationships between scholarship and law that have influenced my perspective on society. Finally, I am grateful to my parents who have always supported my educational aspirations and continue to bless my life with their love and presence. CHAPTER I INTRODUCTION On the evening of Tuesday October 16, 2012 following the second of four televised presidential debates in the 2012 election, Barack Obama's campaign published a post on its official Facebook Page that generated more than 25,000 comments within 48 hours. The post consisted of a photo of the debate under a text caption that stated, "Team Obama had a big win in the second debate-and it's because the President has the right plan to move us forward: http://OFA.BO/e4Q94o." The tens of thousands of comments posted by Facebook users represented a diversity of opinions concerning the debate, the candidates, and various other political issues. Although these comments varied widely in length, argumentative skill, language, and political partisanship, it was clear that this digital platform had cultivated a vast space for public deliberation of political issues. This study examines democratic deliberation on political candidates online social network sites during the 2012 United States election. Since their emergence during the 2008 presidential primaries, campaign social network sites have developed significantly as they were fully embraced by people and politicians on all levels of government in the 2012 election. As suggested in the opening anecdote, social network sites such as Facebook existed as prominent venues for citizens to engage in deliberative discourse as the public accessibility of these websites attracted the attention of millions of Americans. 2 Despite the capacity of social network sites to intrigue and challenge citizens, campaign professionals, and scholars, the precise role of this digital technology in democratic deliberative processes has yet to be articulated in scholarship. Past theoretical and empirical research has examined the concept of democratic deliberation in face t face, mass media, and digital contexts, but scholars have yet to consider the distinctly different communicative space generated by social network site technology. Technological innovation has undoubtedly altered the structures and forms of democratic deliberation, and the technical and cultural characteristics of social network sites lend to novel discursive conditions that warrant examination. Background to the Study Arguments that support and refute the democratizing influence of Internet communications are not new to scholarship. For example, in his 1994 address to participants of the World Telecommunication Development Conference in Buenos Aires, United States Vice President Al Gore proposed that the Internet would "promote the functioning of democracy by greatly enhancing the participation of citizens in decision-making" (Gore, 1996). In the mid 1990s scholarly publications reflected a similar utopian vision of the Internet as researchers such as Berman and Weitzner (1997) claimed, "The Internet presents us with an opportunity to support the highest goals of democracy. We ought to embrace the Internet and support its continued and growing use in political life" (p. 1319). In particular, deliberative initiatives have been identified as a democratic activity that could be supported and enhanced by Internet technology (Delli Carpini, Cook, & Jacobs, 2004; Price, Nir, & Cappella, 2002). Undoubtedly, utopian visions of the Internet have received a great deal of criticism; however, much of this criticism has 3 focused on identifying factors that hinder digital deliberation with the aim of finding ways to support more successful deliberative conditions (Dahlberg, 2011). Since the late 1990s, digital venues dedicated to political discussions and argumentation have proliferated exponentially as there are literally thousands of sites related to politics on local, national, and global levels (Dahlgreen, 2005). Frequently occurring elections help fuel the amount of political discussion on the Internet as political candidates and campaign professionals continue to execute complex digital strategies to engage voters in online communication. In comparison to predigital campaigning that largely limited campaigning efforts to unidirectional campaign to citizen communication disseminated via outlets such as broadcast television, Internet technology allows for interactive bidirectional communication among citizens and campaigns. In 2012, it was not uncommon for campaigns to employ an array of interactive web platforms and social network sites such as Facebook, Twitter, blog network Tumblr, photosharing site Instagram, and other niche web venues like Pinterest to engage voters (Bykowicz, 2012). A study conducted by Pew Research during the 2012 election concluded that 39% of all American adults had engaged in civic or political activities using social media (Rainie, Smith, Schlozman, Brady, & Verba, 2012). Examples of social media activities reported in this research included posting thoughts about civic and political issues, encouraging others to act on issues and vote, and belonging to political or social groups that were working to advance a cause. In contrast to early online deliberative initiatives that suffered from lack of participation, the widespread use of social network sites by political campaigns and the broad discursive participation by citizens on these sites creates a new context for exploration (Dahlberg, 2001a). 4 Genesis and Justification of the Problem Theoretical development is sought by democratic deliberation scholars to inform the validity of current assumptions and to expand coverage to encompass the new digital contexts in which deliberative initiatives are occurring. There have been calls in this field for more empirical studies, and for these studies to be linked more explicitly with deliberative theories (Delli Carpini et al., 2004; Ryfe, 2005; Thompson, 2008). Additionally, scholars have identified the serious need for examination of democratic deliberation in organic settings (as opposed to experimental conditions) where actual initiatives occur (Ryfe, 2005). Researchers encourage the advancement of qualitative case studies to explore these real life contexts and to probe the underlying deliberative processes at work (Barabas, 2004; Delli Carpini et al., 2004). The study in this dissertation addresses all of these calls through a creative exploration of a largely unexamined digital setting. Findings from this research provide a valuable contribution to the theory of democratic deliberation by providing sound empirical evidence generated from a qualitative analysis of a real case study of deliberative discourse carried out on two political campaigns' social network sites during the 2012 election. According to Delli Carpini et al. (2004), the impact of deliberative discursive politics is highly context dependent. While democratic deliberation has been studied in a variety of political communicative contexts, the newness of social network site technology and the time sensitive nature of electoral politics have prevented comprehensive examination of this phenomenon. Consequentially, the present study is significant because it constitutes a starting place from which to understand the phenomenon of deliberative discourse on campaign social network sites. Further, through 5 investigating the relationship between the campaign and the discursive content on their social network sites, this study has implications for understanding how digital deliberation influences political agendas and official decision making processes. Statement of the Problem The purpose of this study is to understand the phenomenon of democratic deliberation on the political campaigns social network sites of one congressional race in Utah. This research seeks to explore experiences of individual citizens, campaign staff members, and electoral candidates who have engaged in this type of communication in order to provide an understanding of the collective experiences of this phenomenon. In addition to examining discursive exchanges, this study focuses on understanding the relationship between political campaigns and the discourse conducted on their respective social network sites. The research questions derived from this statement of the problem are as follows: 1. What strategies do citizens use to engage in democratic deliberation on electoral candidates social network sites? 2. What is the relationship between political campaigns and the deliberative discourse on their social network site? Definition of Terms Following the research questions put forth to guide this study it is important to clearly conceptualize the terms that are used frequently and prominently in this research report. 6 Democratic Deliberation An analysis of democratic deliberation scholarship reveals that the term has been conceptualized in a variety of ways. Differences in definitions depend on the particular characteristic(s) that are emphasized or made centralized, as well as the scope of communication the researcher chooses to include. The following definition of democratic deliberation used in this study reflects a combination of definitions from several scholars. Democratic deliberation: public citizen discourse, in the form of formal and informal exchanges, that address issues of public concern, where disagreement exists and a collective decision is needed. Public citizen discourse is conceived as a form of participation that is open to the public that emphasizes talking exchanges, in comparison to other types of participatory activities such as voting, volunteering, and lobbying (Chambers, 2003; Delli Carpini et al., 2004). Communicative exchanges are not limited to the face to face format, rather public citizen discourse in modern democracies should reflect the significant influence of mass media and recognize that deliberation can occur through a variety of media including phone conversations, email exchanges, and internet forums (Delli Carpini et al., 2004; Page, 1996). By formal and informal exchanges it is recognized that deliberation can occur under flexible conditions. Unlike other scholars who situate deliberative democracy in the context of idealized formal procedural processes (Dewey, 1954; Gastil, 2000), the definition used in the present study embraces the more flexible type of asynchronous and spontaneous communication that is conducted on digital public forums. This definition also states that deliberation must focus on issues of public concern. To clarify, conversations that are personal in nature and unrelated to issues of broader public issues 7 are not considered (Delli Carpini et al., 2004). The last part of the definition contends that democratic deliberation must involve discourse where disagreement exists and a collective decision is needed. According to Thompson (2008), deliberation cannot exist under conditions where participants are like minded or hold similar views before entering a discussion. Rather, "Some basic disagreement is necessary to create the problem that deliberative democracy is intended to solve" (2008, p. 502). Finally, democratic deliberation must be part of a process for arriving at a collective decision, whether all deliberators agree or not (Thompson, 2008). Such collective decisions are sought in formal election processes when citizens are enlisted to produce a collective decision through the process of voting. Political Campaign Social Network Sites The term political campaign social network sites refers to particular websites that are operated by official political campaigns that represent electoral candidates. Although structural variations related to visibility and access vary from site to site, social network sites are: Web-based services that allow individuals to (1) construct a public or semi-public profile within a bounded system, (2) articulate a list of other users with whom they share a connection, and (3) view and traverse their list of connections and those made by others within the system. (Boyd & Ellison, 2007, p. 211) Social network sites are supported by Web 2.0 technology that allows Internet users to engage in interactive participation and shared content creation (Lilleker & Jackson, 2011). MySpace, Facebook, and Twitter are examples of social network platforms that have generated mass popularity. 8 On public social network sites, individuals associated with official political campaigns may set up accounts, which enable them to operate profiles within a social network site. As the administrator of an account, political campaigns can manage content that appears on their site within the technical infrastructure parameters set by the social network platform. Organization of Dissertation This dissertation is organized into five chapters. Chapter 1 has introduced the project and provided the relevant context for the study by providing background information, the genesis and justification for the problem, a problem statement that articulates two research questions, and definitions of important terms. Chapter 2 provides a complete, yet narrowed survey of literature that is necessary for providing a theoretical foundation for this study. Within this chapter, three areas of literature are addressed. First, democratic deliberation literature is reviewed as knowledge of existing theoretical assumptions and empirical studies serve as a foundation for analyzing the precise deliberative discourse examined in the present study. Second, digital democracy scholarship is surveyed that specifically relates to deliberation. Digital democracy has been used to inform a variety of scholarly contexts, but this section focuses on reviewing arguments concerning its relationship to supporting deliberative scenarios. The last section of the literature review surveys existing research regarding the use of social network sites in elections. This final section is important because it provides context for understanding the digital environment and the political parameters from which data in this study were collected and analyzed. 9 Chapter 3 provides a detailed explanation of the qualitative case study research design that was employed to investigate digital democratic deliberation on political campaign social network sites. This section begins with a description of the phenomenological theoretical frame that guided the method, followed by an explanation of the precise data that were collected from two sources: interviews with campaign staff and multimedia content published on campaigns' social network sites. In addition to justifying this particular selection of data, background information will be presented about the case study that provides the necessary context for understanding subsequent analysis. This chapter will conclude with a complete explanation of procedures used for collecting and analyzing data. In Chapter 4, the results of this study are presented. Using the research questions posed previously to organize findings, the researcher describes the nature of social network site deliberation examined in this case study. Specifically, results address strategies used by citizens to engage in democratic deliberation on electoral candidates social network sites. Additionally, findings regarding the relationship between political campaigns and the deliberative discourse published on their social network sites are presented. Chapter 5 presents conclusions concerning the phenomenon investigated in this research. This section begins by summarizing the study and the major findings that emerged from data analysis. Then the researcher provides an interpretation of these findings to clearly articulate their significance to the phenomenon of democratic deliberation and political communication scholarship more broadly. This interpretation transcends specific results produced by this study in an effort to write more generally 10 about the theoretical and empirical contributions of this body of research. Next limitations of this study are addressed by acknowledging some of the shortcomings that emerged as a result of the research design employed. Finally, suggestions for future research that will continue to improve understanding of the phenomenon are presented. Since this study was exploratory in nature, the findings produced a solid foundation for many new research questions to be investigated. CHAPTER 2 LITERATURE REVIEW The research questions posed for this study touch on three primary areas of scholarship: literature on democratic deliberation, digital democracy and deliberation literature, and literature on the use of social network sites in political campaigning. Each of these three categories is vast, and therefore a comprehensive review for each area is not feasible, nor would it be useful to informing the present research. Consequentially, this chapter will address each category of literature by providing synopses of past findings that are most relevant to this study. Democratic Deliberation In democratic deliberation literature scholars have focused on examining the value of deliberative processes in terms of their contributions to the vitality of a democratic government. Theoretical research largely suggests that democratic deliberation performed under certain conditions generates beneficial outcomes for political systems and society more broadly. Empirical research has lagged behind theory and practice, but a body of studies has emerged that reveals mixed support for optimistic theoretical assumptions. This section will begin by presenting an analysis of claims and conclusions that suggest that democratic deliberation is a beneficial process. Then findings that offer skeptical evaluations of democratic deliberation will be discussed. 12 Strengths of Democratic Deliberation The capacity for democratic deliberation to involve collective decision making is valued as a major strength of this process (Delli Carpini et al., 2004; Gastil, 2000; Gutman & Thompson, 1996; Thompson, 2008). In comparison to individual decision making, collective decisions are presumably superior because a collection of voices brings more information to bear, consequentially resulting in a more informed decision outcome (Delli Carpini et al., 2004). Processes of collective decision making have been shown to increase civic engagement as more people are allowed and encouraged to participate, which can ultimately stimulate more public spirited attitudes (Chambers, 2003; Gutmann & Thompson, 1996; Mendelberg, 2002). Researchers have demonstrated that working as a collective can also increase empathy between citizens (Barabas, 2004; Chambers, 2003; Gutmann & Thompson, 1996; Mendelberg, 2002). According to Mendelberg (2002), "deliberation is expected to lead to empathy with the other… through an egalitarian, open minded and reciprocal process of reasoned argumentation" (p. 320). Additionally, under deliberative conditions citizens can become enlightened of their own needs, as well as the needs and experiences of others (Mendelberg, 2002). In addition to generating awareness and understanding of opposing perspectives and rationales, researchers suggest that ideal conditions of deliberation can also promote tolerance between diverse groups (Chambers 2003; Mutz, 2006). In comparison to other types of social interaction, empirical research has concluded that deliberative interactions produce different outcomes (Schneiderhan & Kahn, 2008). When comparing deliberative groups to other decision making conditions 13 such as individual thinking and informal discussion, Schneiderhan and Kahn reported that participants who deliberated were more likely to change their opinions. More specifically, individuals that were encouraged to participate in discussion and provide reasons for their opinions were more likely to shift their positions. It is important to emphasize that evidence of attitude change was discovered on an individual level opposed to an aggregate level change, thus noting this important distinction. Additionally, Schneiderhan and Kahn identified inclusion in deliberation as a central mechanism to the deliberative process and they suggested that future research should concentrate on further investigation of this variable. Neblo, Esterling, Kennedy, Lazer, and Sokhey (2010) confirmed the importance of promoting inclusion in deliberative processes concluding that American's willingness to deliberate was more widespread than expected. Using two large national samples Neblo et al. examined the constructs of inclusion and apathy-two variables cited in previous literature as demobilizing to deliberation-by comparing individual's hypothetical willingness to deliberate and their actual participation when invited to engage in a deliberative initiative. Results revealed that willingness to deliberate was much higher than research on political behavior suggests, and that the people most willing to deliberate were actually those who were turned off by standard partisan and interest group politics. "Far from rendering deliberative democratic reforms ridiculous… findings suggest[ed] that the deliberative approach represents opportunities for practical reform quite congruent with the aspirations of normative political theorists and average citizens" (Neblo et al., 2010, p. 582). From their findings, Neblo et al. suggested that 14 integrating deliberative frameworks into more standard forms of democratic participation in society could increase inclusion in deliberative participation. While most researchers typically acknowledge the potential benefits of democratic deliberation presented thus far, some scholars argue that theoretical assumptions lack practicality in realistic settings due to the difficulty in cultivating and maintaining deliberative conditions (Ryfe, 2005). It has been argued that civic forums that are dedicated to democratic deliberation occur too infrequently and are too uncommon to be considered politically significant (Delli Carpini et al., 2004). In his early work Habermas (1989) articulated the impracticality of democratic deliberation in modern democracies. Drawing on ideals from ancient Greek democracy, Habermas' theory of public deliberation envisioned a "public sphere" that could serve as a social space through which citizens and political elites could engage in rational critical debate of political arguments that could become focused opinion to be used in procedural decision making. However, Habermas identified a number of cultural elements in modern society such as consumer economics, public relations, and mass media that effectively prevent citizens from engaging in the type of ideal deliberative discourse that was achieved in classic Greek politics. According to Habermas, the ideological template of the Hellenic public sphere "has preserved continuity over the centuries-on the level of intellectual history," but its realization in contemporary democracies is severely limited by modern culture (1989, p. 4). In his later work Habermas (1996) responded to his own cultural criticism suggesting that civil society and digital technology may offer some reprise for cultivating successful democratic deliberative conditions. Like Habermas (1996), other more recent 15 researchers have not been preoccupied with theorizing or conducting experiments under ideal democratic deliberation conditions. Rather, scholarship reflects a desire to examine new environments such as Internet forums that appear to foster deliberative democracy, with an emphasis toward understanding how contextual factors can positively and negatively effect public deliberation (Delli Carpini et al., 2004). To illustrate this point it is useful to consider Fishkin's (1995) use of the term "incompleteness" to describe the practical shortcomings of ideal discursive conditions. Fishkin recognized that when some citizens are unable or unwilling to weigh in on arguments in a given debate, discourse becomes less deliberative because it is incomplete. According to Fishkin (1995), "in practical contexts a great deal of incompleteness must be tolerated. Hence, when we talk of improving deliberation, it is a matter of improving the completeness of the debate and the public's engagement in it, not a matter of perfecting it" (p. 41). Thus far, evidence has been presented that democratic deliberation has beneficial consequences on society because it can lead to increased civic engagement, empathy and understanding of opposing viewpoints, and tolerance of diverse beliefs among various social groups. Additionally, it has been demonstrated that empirical research can provide useful suggestions for improvement and practical reform of deliberative practices in realistic settings. But as acknowledged previously, democratic deliberation research has received a fair amount criticism surrounding its usefulness and practicality in contemporary society. Criticisms of Democratic Democracy A summary of past research reveals that deliberation executed under less than optimal circumstances can be ineffective and even counterproductive and dangerous to 16 democracy. For example, Mutz (2006) analyzed surveys and experiments to develop an argument that exposure to oppositional viewpoints during deliberative discursive encounters increased deliberators' ambivalence in a manner that stifled political participation. Survey results showed that most citizens chose to avoid social conflict when discussing politics and consequentially, exposure to disagreement in deliberative settings reduced people's motivations to participate in political activities such as voting (Mutz, 2006). Lack of inclusion and representation among participants in deliberative initiatives has also been cited as a major flaw in democratic deliberation research. Scholars claim that deliberative settings continue to be unrepresentative because they remain populated by the same group of affluent Americans (Delli Carpini et al., 2004). According to Ryfe (2005), allowing greater citizen input in policy making processes would provide a simple solution to this problem, but ultimately this is not viable because people simply do not want to participate. Ryfe explained that citizens' aversions to participating were stimulated by several issues such as feelings of cynicism toward the government, preferences for cognitive heuristics, and desires to avoid responsibility for decision making outcomes. Similarly, Sanders (1997) argued that deliberation does not appeal to "ordinary citizens, or at least not to many residents of the United States, at least not given the way we live now" (p. 348). This realization significantly contributes to unrepresentative deliberative conditions as citizens refrain from opportunities to participate in deliberation. Sanders (1997) publication titled "Against Deliberation" explored several arguments that challenged idealistic notions of deliberation as she explained why 17 "[deliberation] should not necessarily and automatically appeal to democratic theorists" (p. 348). She claimed that it was not possible for society to meet the conditions required for deliberation, which include achieving mutual respect, equality in resources, and guarantee of equal opportunity to articulate persuasive arguments. To elaborate, Sanders wrote: The material prerequisites for deliberation are unequally distributed... because some Americans are more likely to be persuasive than others, this is, to be learned and practiced in making arguments that would be recognized by others as reasonable ones-no matter how worthy or true their presentations actually are. It is also because some Americans are apparently less likely than others to be listened to; even when their arguments are stated according to conventions of reason. (Sanders, 1997, p. 349) Sander's argument revealed a significant threat to democratic ideals as unrepresentative deliberation can clearly subject citizens to conscious or unconscious manipulation and bias in information consumption (Delli Carpini et al., 2004; Sanders, 1997). The capacity for unequal social conditions to allow certain arguments to be embraced while others are dismissed exists as a dangerous consequence of democratic deliberation. A final problem that is consistently cited in the literature as limiting the effectiveness of democratic deliberation lies in the fact that deliberative processes are often disconnected from actual decision making (Delli Carpini et al., 2004; Ryfe, 2005). As stated by Delli Carpini et al. (2004), critics complain that "civic forums are ‘just talk'-idle chat that is cut off from government decision making about important issues" (p. 321). The importance of this issue was elucidated by Ryfe (2005) who argued that the realization of deliberative democracy was dependent in part on successful linking of deliberative practices to policy decisions made in a political system. Ryfe analyzed three 18 reasons for the disconnect between deliberation and policy making. First, deliberative initiatives may be explicitly designed to avoid linkage between deliberation and policy making. For example, rather then trying to influence policy, initiatives such as National Issue Forums (NIF) focus on achieving educational objectives through learning and reflection. Second, deliberation may serve a consulting function as in cases where representative bodies are mandated to call on citizens to share their opinions through deliberative practices such as deliberative polls and citizens juries. In such instances deliberative outcomes can serve as a useful gauge of public opinion, however representatives are not required to abide by them in actual policy making processes. Lastly, when making formal decisions officials do not consider the views of the "ordinary citizen" because they are explicitly bound to drawing on deliberation from relevant stakeholders in a defined systematic process. According to Ryfe (2005), deliberative efforts that are not incorporated in policy making represent a "structural ambivalence within deliberative democracy about the relationship between talk and action" (p. 61). Further, "It is one thing to argue abstractly that contemporary politics might be reinvigorated with greater deliberation and participation. It is quite another to make interactions between ordinary people and policy makers actually work" (2005, p. 62). As acknowledged previously, empirical work has trailed behind theoretical publication and the field of democratic deliberation suffers from failure to explicitly tie empirical research to deliberative theories (Delli Carpini et al., 2004; Ryfe, 2005; Thompson, 2008). Delli Carpini et al. (2004) suggested that future research should explore Americans current political behaviors to discover what deliberative experiences are actually like. Using multiple methods such as qualitative case studies, participant 19 observation, and other field based research, new studies should attempt to "observe more real-world deliberative experiments that occur every day" (2004, p. 336). To further connect empirical findings to theory, Thompson (2008) recommended that future research investigate deliberation among citizens, as well as between citizens and their representatives. The present research addresses all of these suggestions for improvement, as this study contributes valuable empirical findings to this scholarly field. Digital Democracy and Deliberation Literature that examines deliberative democracy in digital environments seeks to understand the extent and quality of critical rational debate using Internet communications, as well as to identify factors that facilitate and hinder deliberative procedures and outcomes (Dahlberg, 2011). Proponents of digital democratic deliberation have suggested that digital technology can offer an important source for information and opportunity to extend the role of the public in the political arena (Hague & Loader, 1999; Papacharissi, 2002). Other scholars remain skeptical of the use of the Internet for practicing democratic deliberation and have warned of the dangers of polarizing effects that can potentially create alienation and disillusionment among citizens (Habermas, 1996; Sunstein, 2001). This section will review digital democracy literature, which seeks to examine the relationship between the Internet and democratic values, as it relates to online deliberative initiatives. Positive Impacts of Digital Technology In the early 1990s, discourse surrounding the Internet was infused with notions of optimism and hope that this new technology would somehow have a positive impact on 20 the depressing health of democracy (Dahlgreen, 2005). The emergence of Internet communication technologies (ICT's) provided opportunities to rethink and possibly replace the institutions, actors, and practices that were thought to contribute to frail conditions of democracy and poor public regard for the existing government (Hague & Loader, 1999). In comparison to traditional mass media, scholars pointed to characteristics of the Internet such as the endless availability of space and ideological breadth, which could ideally support the emergence of an impressive diversity of opinions. According to Delli Carpini et al. (2004), a review of early digital deliberation research reveals that the Internet once offered a solution to the challenge of "durability of changes in attitudes, opinions, and knowledge, and the practicality of the design as a means of increasing meaningful deliberation among the larger population" (p. 334). Additionally, critical scholars have enthusiastically identified the potential of the Internet to allow marginalized social groups that were traditionally excluded from mainstream media and political debate to use the Internet to gain exposure and to advance their interests (Downey & Fenton, 2003; Kellner, 1999). Researchers have argued that the Internet can reduce many practical challenges that inhibit traditional, nondigital democratic deliberative formats. For example, use of the Internet as a communication forum enables more long term deliberation (Delli Carpini et al., 2004). Rather than being limited to strict time and place boundaries, online discussion groups allow Internet users to contribute to debates and consider others' arguments at their leisure (Tambini, 1999). Additionally, Kolb (1996) concluded that asynchronous computer mediated communication was conducive to deliberative conditions because it encouraged reflexivity and justification for speech claims. He stated, 21 [T]he rhythm of e-mail and mailing list exchange encourages opposing manifestos and summaries but also quick movement from what you just said to the arguments and presuppositions behind it. Positions get examined from a variety of angles, and there will be demand for backing on specific points. (Kolb, 1996, pp. 15-16) While Kolb (1996) addressed email and listserv formats specifically, this quote remains significant to the present study because the same type of asynchronous communication is ubiquitous in newer online communication forums such as social network sites. Price, Cappella, and Nir provided empirical support for the strengths of online deliberation through a series of publications that examined the effects of an online deliberative initiative called The Electronic Dialogue Project (Cappella et al., 2002; Price & Cappella, 2002; Price et al., 2002; Price, Goldwaite, Cappella & Romantan, 2003). Data for their analysis were extracted from a multiwave survey that was gathered from a yearlong panel study. Using a random national sample, the project involved the organization of 60 groups that engaged in a series of monthly, real time electronic discussions about issues facing the country and the unfolding 2000 presidential campaign. According to results, online deliberation fostered increased political engagement and general communication participation (Price & Cappella, 2002). Additionally, researchers reported that encountering disagreement in political conversations contributed to individuals' abilities to ground their viewpoints in supportive arguments, and to understand opposing arguments (Price, Cappella, & Nir, 2002). As mentioned previously, broad accessibility of the Internet is conducive to democratic deliberation because it significantly increases opportunities for participation that are not possible in face to face or broadcast venues. Early research suggested that cost elements associated with Internet use were problematic as users had to incur three 22 significant costs including: the price of a personal computer or other hardware, the cost of Internet services, and finally telecommunications services that connected users to the online service (Berman & Weitzner, 1997). However, the availability of more affordable access options have allowed the low cost of Internet use to be advantageous for both organizers and participants in democratic deliberative scenarios. To emphasize, Pew Research conducted a national survey of Internet use at the end of 2009 and found that 74% of American adults (ages 18 and older) use the Internet (Rainie, 2010). In addition to cost, the open architecture of the Internet has been praised for supporting democratic conditions by offering greater accessibility to information and overcoming problems of space scarcity that are relevant to print and broadcast media (Tambini, 1999). Because the web is a decentralized network, "Anyone with content to publish or ideas to exchange can do so from any point in the network," consequentially supporting a great diversity of opinions, ideas, and information (Berman & Witzner, 1997, p. 1314). According to Wilhelm (1999), democracy means inclusiveness, ensuring that all people who are affected by a policy have the opportunity to deliberate. Digital technology reinvigorates the possibility for inclusiveness, thus creating a greater availability of conflicting views to be made available for public consideration. Problematic Issues Related to Digital Technology Despite optimism that the open and decentralized infrastructure of the Internet leads to increased opportunities for participation in deliberative initiatives, technology alone cannot guarantee that people will be motivated to take part in the type of engaged argumentation among opposing opinions that is required for successful democratic deliberation (Bohman, 1998; Thompson, 2008). For example, Dahlberg (2001a) 23 examined a particular listserv that was constructed explicitly for the purposes of fostering deliberation. He concluded that while successful deliberation could be engaged through the listserv, its vitality was highly threatened by lack of participation. Dahlberg claimed that ultimately, citizens were more attracted to the popularity of commercial sites, as well as virtual communities of common interest where participants could engage in communication with like minded others. This finding supports literature that suggests that citizens prefer to avoid social conflict that arises when discussing political issues with others who hold opposing views (Mutz, 2006). When presented with increased opportunities to deliberate and be included in broader collective decision making, citizens continue to limit their exposure to discussions with like minded individuals, thus cultivating the existence of polarized opinion groups (Avery, Ellis, & Glover, 1978). Habermas (1998) considered this tension between increased polarization and augmented inclusion in his discussion of the impact of new ICTs on public deliberation. While he acknowledged that the growth of technology increases the amount of information available to a wider public, he also expressed major concern for the fact that high pluralism in Internet audiences does not necessarily expand "intersubjectivity" or "discursive interweaving of conceptions" (1998, p. 120). Rather, to illustrate fragmentation in online populations, Habermas stated, "The publics produced by the Internet remain closed off from one another like global villages" (1998, p. 121). Dahlberg (2001a) used the term "mutually-exclusive cyber-communities" to describe this same polarizing phenomenon (p. 618). He wrote, While a great diversity of communication takes place across cyberspace, some of which does involve critical discussion of controversial issues, many participants simply seek out groups of like-minded others where member's interests, values and prejudices are reinforced rather than challenged. (Dahlberg, 2001a, p. 618) 24 For more than a decade, scholars have agreed that this issue of audience fragmentation exists as a significant threat to deliberative democracy (Graber, 2003; Howard, 2005; Sunstein, 2001). Sunstein (2001) offered a particularly pessimistic outlook for the future of digital deliberative initiatives. He claimed that the infrastructure and growth of the Internet produces social enclaves composed of very like minded content and discussion groups, which ultimately widen gulfs between extreme sides on public issues. More specifically, Sunstein identified two preconditions of democratic deliberation that are threatened by the Internet: 1) inadvertent exposure to diverse viewpoints, and 2) sharing of common experiences. In regard to the first condition, the Internet enables people to filter information resulting in restricted exposure to diverse perspectives. Howard's (2005) study of digital campaign strategy provided empirical evidence of filtering effects. Through observation of extensive information filtering carried out by campaigns in their efforts to customize digital messages, Howard concluded that the quality and quantity of shared text that could be accessed by the public was severely reduced. With reference to Sunstein's (2001) second precondition of a healthy democracy, the sharing of common experiences is also threatened by the Internet because this technology allows people to isolate themselves to only hear particular vantage points. Self isolating makes it difficult for people to understand and solve collective problems that a heterogeneous society faces together. Contrasting findings regarding the amount of Internet fragmentation existing on the Internet during the 2004 election were reported in a study by the Pew Internet & American Life Project and the University of Michigan School of Information (Horrigan, Garrett, & Resnick, 2004). Using a survey that assessed respondents' exposures to 25 various political arguments, as well as their Internet use, media use, interest in the campaign, and open mindedness, Horrigan et al. concluded that the Internet contributes to a wider awareness of political arguments. Rather than limiting their exposure to opposing opinions, Internet users "are more aware than nonInternet users of all kinds of arguments, even those that challenge their preferred candidates and issue positions" (2004, p. ii). A more recent Pew study provided further evidence that the Internet and social network site users in particular are frequently exposed to diverse and opposing political opinions (Rainie & Smith, 2012a). According to results from a survey conducted in early 2012, 73% of social network site users only sometimes agreed or never agreed with their friends' political postings. Additionally, they found that 28% of users usually responded to opposing viewpoints with comments or posts of their own. This behavior was especially true for citizens categorized in the most politically engaged demographic who reported that they challenged friends' social network site material about politics if they disagreed with it (Rainie & Smith, 2012b). However, this same study discovered that some social network site users employed political material that was posted on the sites to assess the vitality of their online relationships. More specifically, 18% of respondents reported shunning people in their online network for reasons such as posting something about politics that they disagreed with or found offensive. While it appears that the Internet and social network sites can facilitate exposure to divergent political perspectives, users also actively avoid or block the appearance of conflicting opinions in their online interactions. A review of past literature demonstrates that issues regarding fragmentation, polarization, and filtering of information appear to be major concerns for research in 26 digital democratic deliberation. Additional problems identified by Dahlberg (2001b) include: increasing colonization of cyberspace by state and corporate interests, a deficit of reflexivity, a lack of respectful listening to others, the difficulty of verifying identity claims and information put forward, the exclusion of citizens from online political forums, and the domination of discourse by certain individuals and groups. Among his suggestions for improving deliberative outcomes on digital forums, Dahlberg suggested that new models and technologies should aim to increase audiences by offering, "seductive and easily consumable options" for deliberative engagement that would attract audiences that were previously hostile towards public deliberation (2001a, p. 629). While social network sites-the digital format under scrutiny in the present study-surely do not operate solely to stimulate a virtual deliberative space, this research attempts to explore the deliberative behavior of people using this popular and publicly accessible communication technology. Clearly, past literature has generated mixed findings regarding the capacity of the Internet to foster a space for successful democratic deliberation. However, it is important to emphasize that digital formats vary considerably in technical and cultural structure that may significantly impact the success of deliberative activity. For example in research by Tambini (1999), multiple online civic network forums that varied in terms of their deliberative conditions were examined. Following analysis it was concluded that the vitality of deliberation was significantly influenced by contextual variables such as access and motivation. As a result of evolving digital communication there is a continuing need to assess the potential for newer technologies (such as social network sites) to cultivate successful deliberative conditions. 27 The next section will discuss characteristics of social network sites that make them distinct from other Internet technologies. Where other digital technologies fall short of fostering a deliberative atmosphere, the distinct design and cultural structure of social network sites may offer solutions that can support and even strengthen deliberative conditions. Social Network Sites in Political Campaigning The use of social network sites in elections has become commonplace for politicians as this technology has become integrated in communication media for campaigns on all levels of government. Scholars began exploring this topic during the 2006 election and the majority of this literature has concentrated on describing how social network site technology was used by campaigns (especially presidential) and whether it generated positive or negative outcomes for campaigning strategy. For example, researchers have focused on evaluating the success of social network sites in public relations efforts (Cogburn & Espinoza-Vasquez, 2011; Metzgar & Maruggi, 2009). Others have examined the technical and cultural infrastructure of social network sites to comment on conventional political communication constructs such as political participation (Boyd, 2008; Geuorgieva, 2009) and political cynicism (Hanson, Haridakis, Cunningham, Sharma, & Ponder, 2010). Based on this review of literature there is a noticeable absence of scholarship that specifically examines democratic deliberation on campaign's social network sites. This section provides an overview of what existing literature concludes about campaign social network sites more generally in order to provide a context for understanding their potential for engendering deliberative activity that is examined in the present study. 28 Technical and Communication Characteristics of Social Network Sites Although structural variations concerning visibility and access differentiate individual social network sites, they can essentially be conceptualized as Internet platforms that allow individuals to construct profiles within a bounded system, articulate lists of other users with whom they share a connection, and send public or semipublic messages (Boyd & Ellison, 2007). In terms of their technical structure, social network sites offer politicians the opportunity to connect with citizens 24 hours a day and seven days a week through a method that is both inexpensive and efficient. For citizens, social network sites create opportunities to practice civic skills such as joining a political group or sharing a political link without having to dedicate too much time and effort (Vitak, Zube, Smock, Carr, Ellison, & Lampe, 2011). The infrastructure of social network sites was enabled by Web 2.0 technologies that allow users to generate shared content on the World Wide Web. Unlike earlier applications on the Internet that limit users to consuming information in a more linear fashion, the technology of Web 2.0 fosters the construction of virtual communities inhabited by interactive participants who contribute to the creation of website content (Bruns, 2008; Lilleker & Jackson, 2011). The Internet's reliance on bandwidth enables this media to successfully overcome barriers of geographic space and time that potentially stymie discourse between politicians and constituents. Additionally, beyond accessibility social network sites require no additional costs for communicating with one or one million people. Unlike broadcast television, which largely restricts use to campaigns with sufficient resources, the low cost of production and access associated with social network sites assists to level 29 the playing field for candidates, interest groups, and regular citizens (Johnson & Kaye, 2000). For example, Gueorguieva (2009) claimed that MySpace and YouTube were especially advantageous resources for low budget campaigns and political organizations during the 2006 election cycle because they provided an inexpensive means to collect campaign contributions and organize volunteers. The scale of communication offered by social network sites is especially alluring to politicians who envision the Internet as a democratizing technology (Boyd, 2008). In their research on the 2008 election, Metzgar and Maruggi (2009) concluded that when properly employed, social media could function as a highly relevant and cost effective campaign tool. Online publication of even a very small amount of digital characters could potentially capture a nation as information travels, "from idea, to digital post, to a national audience with very few gatekeepers or powerbrokers weighing in on that information" (2009, p. 151). But while an infinite reach may be structurally possible for public social network site messages, merely publishing content does not guarantee reception (Boyd, 2008; Metzgar & Maruggi, 2009). While social network site technology offers a timely and inexpensive means for political communication to be conducted, messages may not always be used, received or interpreted as expected. When individuals communicate using digital media, a range of cultural representations, experiences, and identities are created (Coleman, 2005). While different populations use social network sites for different reasons, studies have concluded that people primarily use this technology for interpersonal and intergroup purposes, and to keep in touch with friends or to intensify offline relationships (Ellison, Steinfeld, & Lampe, 2007; Hanson et al., 2010). According to Bargh and McKenna (2004), relative 30 anonymity that exists on the Internet makes self disclosure easier, thus facilitating improved relationship formation. In comparison to face to face encounters, online communicators may be more likely to cultivate close and meaningful relationships using the Internet as a mediating device. Recognizing characteristics of social network site relationships is important to understanding how politicians and citizens use this technology to communicate. When confronted by political information on social network sites, social network culture may cause individuals to behave differently than they would in face to face and other mass mediated environments. For example, findings from a study concerning citizens Facebook use during the 2008 election concluded that the norms of political activity on this social network site were nuanced (Vitak et al., 2011). More specifically, participants felt that personal expression of political views was appropriate, but efforts to persuade or recruit others were not. Further investigation of the use of social network sites in electoral politics is necessary in order to more fully understand the impact of this technology in campaign communication. Social Network Sites in Elections Unlike mass communication venues, the interactivity presented by social network sites promise increased engagement and dialogue (Hoffman & Kornweitz, 2011). For political campaigns, this means that in addition to disseminating information and capturing data about citizens, social network sites also offer a mechanism for ongoing engagement (Metzgar & Maruggi, 2009). Unlike some digital technologies that are designed and used primarily for information consumption, social network sites encourage citizens to actively engage, and to "create their own political content, distribute it online, 31 and comment on the content created by others" (Hanson et al., 2010, p. 585). This ability for ordinary citizens to participate in the political realm compromises the traditional top down one way communication style and forces political actors to modify their practices to a bottom up approach. According to Metzgar and Maruggi (2009), "Social media's strength lies in its communal nature and lack of strict hierarchies. Campaigns that embrace this lack of hierarchy, rather than fight it, are more likely to reap the benefits the technology can offer" (p. 161). The disruption of political hierarchy created by social network site technology is also significant in its ability to allow silenced or marginalized voices to emerge, thus transforming the power dynamics of politics by giving voice to the nonpolitically elite (Fraser & Dutta, 2008). Social network sites make it simpler for ordinary citizens to communicate with government officials as users are presented with the opportunity to interject opinions into the political sphere in new ways that did not exist prior to the emergence of this technology. While giving ordinary citizens a voice presents opportunity for lively democracy, Lilleker and Jackson (2011) insist there is danger in allowing the public to lead an agenda that was once carefully controlled by professional campaigners. When campaign content is user generated, candidates lose control and consistency of their image and message (Gueorguieva, 2009). Candidates' fear and resistance to interactive technology was illustrated in Stromer-Galley's (2000) early study of Internet use in the 1996 and 1998 elections. According to analysis of candidate websites and interviews with campaign staff members, Stromer-Galley concluded that the Internet was not being used to promote increased deliberation between citizens and politicians because "campaigns did not want 32 to open up the possibility for burdensome exchange among candidates, campaign staffs, and citizens, which could entail losing control over the communication environment and losing the ability to remain ambiguous in policy positions" (2000, p. 112). In 2012 candidates appear to have fully overcome their early resistance to interactive digital technology. To exemplify this point it is useful to consider Barack Obama's statement for reelection that was announced through the use of multiple interactive media platforms. A year and a half prior to Election Day, Obama disseminated the question "Are you in?" using email, text messaging, Twitter, YouTube, and an app that connects supporters and their Facebook friends to his campaign website (Preston, 2011). During the summer leading up to the 2012 election the Pew Research Center's Project for Excellence in Journalism conducted a study to investigate the presidential candidates' uses of digital technology. In particular, they concentrated on examining the extent to which candidates used social network sites to engage in dialogue with citizens. According to Pew's analysis, social media technology was primarily used for one way messaging rather than interactive communication. "Neither campaign made much use of the social aspect of social media," as the candidates rarely replied to citizen messages or engaged in dialogue other users (Pew, 2012). This result suggests that digital democracy theory, which claims that digital technology will allow political elites to engage in a new level of dynamic conversation with voters, is not being realized in actual practice. Results from a more recent Pew study performed the week prior to 2012 Election Day revealed that social network sites were a significant part of the process by which voters-especially young voters-talked about their ballot selections (Rainie, 2012). More specifically, the study stated, "22% of registered voters have let others know how 33 they voted on a social networking site such as Facebook or Twitter" (2012, p. 2). This result is important because it provides evidence that nearly one fouth of citizens are using social network sites as an outlet for expressing their political opinions. In addition to focusing on the impact of online communication in the digital realm, researchers have also explored the extent to which social network site engagement translates to campaign communication outside the parameters of social network sites (Cogburn & Espinoza, 2011; Erikson, 2008; Geuorguieva, 2009). For example, Geuorguieva (2009) suggested that candidate involvement on the social network site MySpace could act as a catalyst to organize door to door canvassers. Similarly, Cogburn and Espinoza (2011) analyzed Obama's use of social network sites in 2008 and discussed the potential for social network site activity to build a geographically distributed virtual community that could translate to "on-the-ground activism" (p. 192). Cogburn and Espinoza concluded that the presidential candidate's social network sites "extended beyond the campaign offices and allowed staff, volunteers, and the public to stay connected" (p. 200). The dynamic between online social networking and offline social networking is important because it speaks to the potential for digital technology to strengthen democratic ideals through increased political participation in activities such as voting, volunteering, and engaging in other civic events. As demonstrated thus far, the integral relationship between social network sites and political campaigns has stimulated a wide range of implications for democratic processes and campaign communication specifically. Boyd (2008) claimed that social network sites are especially appealing to political candidates because this technology supports the "desire to exhibit oneself for the purposes of mass validation" (2008, p. 241). 34 According to Slotnick (2009), there is a natural connection between social network sites and political campaigns because both are designed to encourage group formation and are dependent upon strong and persistent networking. Regarding Facebook specifically, Slotnick suggested that political campaigns and social network sites maintain a mutually beneficial relationship, as they both rely on constant communication and encourage mobilization of supporters. Existing scholarship demonstrates the popularity of campaign communication on social network sites and also reveals many implications stimulated by this type of digital civic engagement. However, research has yet to focus on social network site deliberation, which is represented in the form of debates and compromises that occur when citizens and policymakers puzzle their way through problems that are crucial to achieving ideals of active political citizenship (Fishkin, 1991; Habermas, 1989). Chapter 3 presents the method used in this research to investigate this line of inquiry for the purposes of further understanding an important type of political communication. CHAPTER 3 RESEARCH DESIGN This study examines the phenomenon of democratic deliberation conducted on political campaigns' social network sites. The research design described in this chapter is grounded in the philosophical assumptions of a transcendental phenomenology. This theoretical frame emerged as the most appropriate qualitative approach because of its perceived capacity to authentically capture the subjective and value laden from a purposeful, nonrepresentative sample. Following a brief description of this theoretical tradition, specifics of the method that were motivated by philosophy and literature related to phenomenology will be articulated. Explicit details will be provided regarding the precise data examined, followed by a description of contextual and background information that is necessary for understanding the data. Lastly, procedures for collecting and analyzing are set forth. It should be emphasized that decisions made in this research design were purposeful and based on the potential for gaining the most comprehensive understanding of the experience of democratic deliberation on social network sites. Phenomenology According to Creswell (2007), "a phenomenological study describes the meaning for several individuals of their lived experiences of a concept or a phenomenon" (p. 57). Phenomenology is used when an understanding of common experiences is needed in 36 order to develop practices or policies. This theoretical framework is appropriate for addressing the research questions proposed in this study. A deep understanding of deliberating on political campaign social network sites is sought so that the experience of democratic deliberative initiatives can be strengthened for future participants. Drawing primarily from Creswell's (2007) and Moustakas' (1994) literature on using phenomenology as a method, the procedure employed in this study involves collecting data from several persons who have lived through and can describe the phenomenon, followed by analysis that will produce a description that conveys the overall essence of the experience. Phenomenologists believe that it is impossible to separate method from philosophy because human experience is formed around the interests and intentions that give it meaning. Consequentially, a philosophical overview is a necessary precursor to explaining the method used in this study. Phenomenological scholarship is a highly varied field of inquiry that reflects the thinking of a number of philosophers including Edmund Husserl, Maurice Merleau-Ponty, Martin Heidegger, and Alfred Schutz (Miller, 2005). Two main approaches to phenomenology include hermeneutic phenomenology and transcendental or psychological phenomenology, the latter being the approach used in the present investigation. As described by Moustakas (1994), a transcendental phenomenology is focused less on researcher interpretations (a feature that is central to a hermeneutical phenomenology), and concentrates more on describing the experiences of participants. The transcendental approach is influenced heavily by the work of Husserl-regarded by Vandenberg (1997) as "the fountainhead of phenomenology in the twentieth century" (p. 11)-who was concerned with foundational issues of how we come to know the world 37 (Miller, 2005). Husserl's (1931, 1973) philosophy opposed empirical science procedures that claimed to produce sovereign truths. Rather, he believed that the foundation for philosophic understanding was discovered through the primacy of lived experience. For Husserl, lived experience (sometimes referred as Lebenswelt or "life world" in phenomenological literature) was the primary context from which all human endeavors- including natural science-take their beginnings and orientation. There are several basic assumptions that are shared by most phenomenologists. First, phenomenology involves the search for the essence, or the central underlying meaning of experience (Creswell, 2007; Moustakas, 1994; Natanson, 1973). Second, there is emphasis on intentionality of consciousness (Stewart & Mickunas, 1990). This concept comes from Husserl's idea that knowledge is not found in external experience, but rather in individual consciousness that is always directed toward an object; and consequentially, meaning of that object is always relative to one's consciousness derived from personal background and current life events. Third, philosophy of phenomenology necessitates that a researcher must "suspend all judgments about what is real-the ‘natural attitude'-until they are founded on a more certain basis" (Creswell, 2007, pp. 58-59). Husserl (1931, 1973) used the term "epoche" to describe the process of bracketing information, or setting aside prior knowledge, as crucial to gaining a fresh perspective toward the phenomenon that is free from interference of interpretive influences. This point is especially relevant to the transcendental approach as hermeneutic and existential phenomenologists argue that it is impossible to completely bracket prior conceptions and knowledge because consciousness cannot be separated from being (Heidegger, 1962). LeVasseur (2002) offered a modification to Husserl's 38 philosophy of bracketing suggesting that this thinking could be understood as an extension of natural curiosity. When we become curious "we no longer assume that we understand fully, and the effect is a questioning of prior knowledge" (2002, p. 417). According to Creswell (2007), identifying and specifying personal philosophical assumptions prior to collecting data allows the focus of their study to be directed toward participants opposed to personal experiences of the researcher. To summarize, phenomenologists attempt to discover the essence of a phenomenon by learning what an experience means for a person through recognition of individual consciousness. From individuals' descriptions, more general or universal descriptions can be derived to capture the essences or structures of an experience (Moustakas, 1994). Data The data in this study were gathered from two different sources: social network sites of two different political campaigns and interviews conducted with people involved with social network site communication for each of the respective campaigns. The campaigns selected for analysis represented opposing candidates that ran in a congressional race in Utah's newly formed 4th District during the 2012 general election. The Democratic candidate, Jim Matheson, ran as a sixth term incumbent from Utah's 2nd District. His digital strategy for communicating campaign information to voters included the use of email updates, Twitter, Facebook, and Youtube. The Republican candidate, Mia Love, was serving her first term as mayor of Saratoga Springs, a position she was elected to in 2009. Her digital strategy consisted of the use of Twitter and Facebook. 39 These two campaigns were selected among other electoral races throughout the United States for three main reasons that included: 1) ideal activity level, 2) potential for comparison, and 3) competitiveness of the race. First, based on preliminary observation of social network sites the two sites selected for analysis reflected frequent and lively critical argumentation of political issues, as opposed to other campaigns' social network sites that appeared to have been neglected by either the campaign staff or citizens. Additionally, there was a manageable amount of content being published on each of these sites that was conducive to the type of in depth qualitative analysis that was necessary for revealing the complexity of the phenomenon (Miles & Huberman, 1994). To clarify this point, an attempt to collect and analyze a massive quantity of multimedia content produced on national campaigns' social network sites would have resulted in data overload (Miles & Huberman, 1994). Trying to process the amount of data produced on sites such as Obama's or Romney's Facebook Pages would have made it difficult to obtain the useful and relevant information needed to answer the inquiries sought in this study. Second, rather than considering just one campaign in an electoral race it was necessary to examine data from the competing candidate. This provided a more complete understanding of what it meant to engage in deliberation concerning this political event. Because phenomenology seeks to understand a more general or universal perspective as opposed to an individual's experience, collecting data from two candidates allowed the researcher to capture a more holistic understanding of this deliberative experience (Moustakas, 1994). By observing text on both campaigns' social network sites it was possible to detect whether similar individuals or groups were engaging on both sites, and 40 how communication differed in terms of deliberative approaches and outcomes. Additionally, looking at opposing sides of the race allowed the researcher to explore whether there was deliberative communication being engaged between the two sites in the form of responding to, or referencing content on the opposing candidate's site. For example, deliberation between the sites could be facilitated by reposting or hyperlinking between the sites. The third reason for selecting the Matheson vs. Love election was because it represented a highly competitive race in which both campaigns carried out aggressive media campaigns, held debates, and received significant coverage in local news outlets. In addition to communication featured on Matheson's and Love's social network sites, letters to the editor published in news outlets, as well as the prevalence of volunteer efforts throughout the precinct provided evidence that this election was being publicly discussed. As indicated by the closeness of preelection polls and the final vote count, this race also stimulated considerable disagreement among the electorate (Romboy, 2012a; Canham, 2012a, 2012b). To summarize, a high level of competition, as indicated through aggressive campaigning, media coverage, public discussion, and public disagreement, constituted a combination of elements that presumably motivated democratic deliberation, thus creating an ideal case study for the current research. Background Information About Data The race between Matheson and Love was the first congressional election for Utah's 4th District that was formed when the Utah Legislature drew new congressional boundaries for the 2012 election. This district contained southwestern Salt Lake County, western Utah County and all of Juab and Sanpete counties. Slightly more than one month 41 prior to Election Day, local and national media referred to this race as "among the most hotly contested congressional fights in the country" (Gehrke, 2012a). Another source suggested that monetary contributions totaled "an unheard of amount of money for a House race in Utah" (Canham, 2012b). According to final totals released after Election Day, Matheson defeated Love by just 768 votes in a competition that was the most expensive House race in Utah history (Canham, 2012a). This section presents background information about the election in an effort to provide context for understanding the phenomenon under investigation. Knowledge about the candidates, as well as the amount of expenditures and donations, issues prevalent in media discourse, and the communication strategies executed by both campaigns are necessary for gaining a comprehensive understanding of the deliberative experience under investigation. Prior to the creation of Utah's 4th District, Jim Matheson had been Utah's only Democratic House member representing Utah's 2nd District for the past six consecutive terms. After redistricting that occurred on December 15, 2011 divided the 2nd District into four parts, Matheson announced he would run in the 4th District where he faced new competitor Mia Love who reportedly "push[ed] him harder than any of his past opponents" (Romboy, 2012a). Matheson, 52 years old at the time of the election, was a sixth generation Utahn whose father had served two terms as governor (Romboy, 2012a). After studying at Harvard and earning an MBA from University California Los Angeles, he worked in the energy industry developing privately owned power plants. After moving back to Utah with his wife and two sons, Matheson served in Congress beginning in 2000 when he was first elected to office. 42 Candidate Mia Love had a very different story to tell in her campaign narrative. Before being elected the first African American female mayor in Utah, 36-year-old Love had been employed as a fitness instructor and also served on the Saratoga Springs City Council. Throughout her campaign she retold the tale of her upbringing as the daughter of Haitian immigrants who had arrived in the United States with just $10 in their pockets (Romboy, 2012a). After graduating with a fine arts degree from the University of Hartford, Love moved to Utah where she met her husband, joined the Church of Jesus Christ of Latter-day Saints (LDS), and had three children. In 2003 she won a seat on the Saratoga Springs City Council and eventually became mayor in 2009. Part of the national attention that was directed toward Love's campaign was stimulated by the historic fact that if she had defeated Matheson she would become the nation's first African American Republican congresswoman. From the start, political partisanship was a prominent issue in campaign media discourse as both candidates fought for the Republican vote in a district that was heavily populated by Republican voters (Romboy, 2011). Love's campaign platform demonstrated her strong partisanship to the Republican Party. She was described by The San Francisco Chronicle as the "darling of tea party Republicans," and was honored with one of few slots to speak at the Republican national convention in August 2012 (Foy, 2012). Following her convention speech, Love received a great deal of support from high profile Republicans including former Secretary of State Condoleezza Rice, Arizona Senator John McCain, vice presidential nominee Paul Ryan, and House Speaker John Boehner (Gehrke, 2012b; Romboy, 2012b). Her campaign focused on depicting close ties 43 to presidential candidate Mitt Romney, which was especially reflected in her television ad campaign that aired during the final weeks of the election (Canham, 2012c). For Matheson, the message of his political bipartisanship was strongly conveyed in his campaign communication. In an interview with The Salt Lake Tribune, Matheson emphasized his independent status and Republican appeal stating, "I've always worked across the aisle and that's the way solutions to complex issues get resolved" (Gehrke, 2012c). In addition to past voting records that revealed Matheson's ability to persuade Utah Republicans to vote for him, his television ads touted his crossover support from prominent Utah Republicans (Gehrke, 2012c; Gehrke, 2012d; Roche, 2012). During debates against Love, Matheson claimed, "he would be the first Democrat to reach across the aisle to help Romney if he wins the election," and consistently stated that he was an independent voice that puts Utah first rather than a party (Burr & Canham, 2012; Gehrke, 2012c). The strength of national attention that this congressional race received was evidenced by the amount of independent outside spending by individuals or organizations. Unlike contributions that were coordinated with the official campaign, this outside spending was neither limited nor regulated. As of October 15, 2012 third party groups had reportedly pumped more than $5.4 million into the race that was split almost evenly in support of the candidates (Canham, 2012e). By Election Day more than $6 million was spent by outside groups and resources that state political parties shifted to Utah's 4th Congressional District (Canham, 2012a). Matheson received contributions from Democrat supporting PACs such as Center Forward and House Majority PAC. Love's campaign was heavily financed by the National Republican Congressional Committee, 44 tea party groups FreedomWorks and Club for Growth, and former Defense Secretary Donald Rumsfeld (Canham, 2012b; Canham, 2012d; Canham, 2012e). Overall both candidates received the bulk of their donations from PACs that helped fund hours and hours of ads that blanketed Utah broadcast channels during this election (Canham, 2012a, 2012d). Local media outlet Deseret News described the ad war between Matheson and Love as "a raging inferno" (Romboy, 2012c) and in a different article wrote, "Both candidates have saturated the airwaves with commercials that make the other look sinister. They and their supporters have poured millions of dollars into one of the nastiest and most expensive ad wars in recent memory" (Romboy, 2012a). In addition to the ad expenditures reported to the Federal Election Commission (FEC), thousands of dollars were also spent on issue oriented ads paid for by political nonprofits that were not required to file with the FEC (Canham, 2012d). Outside of paid media spots, Love and Matheson also received free television access through four separate debates that were hosted by local media outlets between September 26 and October 2 (Gehrke, 2012a). Issues emphasized during these debates included: energy, federal spending cuts, and funding for special education (Foy, 2012; Romboy, 2012c; Gehrke, 2012a; Gehrke, 2012f). Following the debates, an independent poll released by Brigham Young University's (BYU) Center for the Study of Election and Politics illustrated the closeness of the competition as Matheson and Love were tied at 43% with 14% of voters undecided (Gehrke, 2012f; Romboy, 2012a). Both candidates used social network sites Twitter and Facebook during the election. Love began using Facebook to campaign in November 2011 and posted consistently every one to two days until Election Day. Matheson's use of these sites was 45 carried over from his 2010 Congressional campaign. In addition to Twitter and Facebook Matheson also operated a YouTube channel where he posted previously aired television ads and other videos produced by both his campaign as well as voters. Procedures The procedures used in this study were derived from Creswell's (2007) and Moustakas' (1994) literature on using phenomenology as a method. They were designed to achieve what Natanson (1966) claimed to be the central endeavor of phenomenology, "to transcend the natural attitude of daily life in order to render it an object for philosophical scrutiny and in order to describe and account for its essential structure" (p. 3). The following sections will explain the precise procedures used for collecting and analyzing data that allowed the researcher to develop the most comprehensive understanding of the phenomena being investigated. Data Collection As introduced previously, data for this study were drawn from the social network sites of two political campaigns and interviews conducted with people involved with the social network site efforts for each of the respective campaigns. Interviews captured the life experiences of individuals involved with the phenomena and enabled the researcher to generate a comprehensive description of the relationship between each campaign and its social network site. Digital data from social network sites were also collected because this content reflected the way that democratic deliberation was promulgated for public consumption. To clarify, social network sites existed as the means by which geographically distanced citizens communicated with each other and made deliberative 46 discourse available to the public. It is acknowledged that this documented data does not capture the "lived experience" of democratic deliberation in the strictest sense of that term, but given the fact that campaign social network sites are special in their capacity to allow deliberators to remain anonymous (to a certain extent) and to publish content whenever and wherever they please, the collection of documented content is not entirely a distortion of lived experiences. Social network sites. Social network site data collection was limited to content featured on the official Facebook Pages of the two candidates-Jim Matheson and Mia Love-that competed for Utah's 4th District congressional seat. Matheson's campaign Facebook Page (not to be confused with his congressional Facebook Page) was located at the web address: http://www.facebook.com/MathesonUT. Love's Facebook Page was located at http://www.facebook.com/miablove. Facebook was selected instead of other social network site platforms such as Twitter and YouTube because the distinct discourse produced on Facebook was consistent with the researcher's conceptualization of democratic deliberation presented in the Chapter 1. To review, the definition of democratic deliberation was specified as: public citizen discourse, in the form of formal and informal exchanges, that address issues of public concern, where disagreement exists and a collective decision is needed. The first criterion, "public citizen discourse," was met because accessibility and participation on the Facebook Pages was open to the public. Additionally, deliberation literature suggests that past studies have suffered from unrepresentative samples of participants that failed to include all relevant parties (Ryfe, 2005). Consequentially, the researcher purposely selected the largest social network site that has sustained popularity among political campaigns for the longest amount of time in 47 an attempt to study the venue that could potentially involve the greatest amount of participants (Stone, 2009; Williams & Gulati, 2009). The second criterion "formal and informal exchanges" was also met because the technical structure of the Facebook Pages enabled users to exchange messages at their convenience as the forum was open to participation twentyfour hours a day, seven days a week. The third criterion in the definition, "address issues of public concern," was met because the discourse on the Pages largely addressed issues surrounding the election, such as the policy positions of the candidates. This information was of public concern because it assisted citizens in making informed voting decisions. The remaining criterion, "disagreement exists and a collective decision is needed" was also satisfied because the Facebook Pages featured opposing views pertaining to the looming collective decision that called upon citizens of the 4th District of Utah to decide which candidate should represent them in Congress. Other social network sites used by the candidates failed to meet one or more criteria that would allow communication to qualify as deliberative discourse so use of these other sites would have prevented the most in depth phenomenological understanding sought in this study. Ultimately, Facebook existed as the social network site that was the most conducive to the ideal deliberative conditions theorized in past literature and consequentially it was selected for scrutiny. Specific data that were collected for coding included all posts and comments featured on both Matheson's and Love's campaign Pages during the month leading up to Election Day-between 12:00 am on October 6, 2012 and 11:59 pm on November 6, 2012. A total of 6,671 comments were collected from 132 posts. More specifically, 1,135 48 comments were collected from 72 posts on Matheson's Facebook Page and 5,536 comments were collected from 58 posts on Love's Facebook Page. To explain what is meant by "posts" and "comments" it is necessary to briefly describe the technical and aesthetic infrastructure of Facebook at the time when these data were collected. The Page format set by Facebook enabled the structure and design of both candidates' Pages to be identical. From top to bottom, each Page featured a "cover photo," followed by a "profile picture" and basic information about the candidate. Below these elements was the "timeline," described by the Facebook Help Page as "a collection of the photos, stories, and experiences that tell your story" (see Figure 1 and 2). On the timeline, Page administrators could publish "posts" that included text, images, or multimedia, and then visitors could publish "comments" under any given post (See Figure 3). The timeline organized posts in chronological order with the most recent updates at the top of the page. This organizational design allowed audiences to view a running dialogue of posts and their related comments that were published. Constrictive time parameters for collection were established in an effort to collect a manageable amount of data that would allow for a thorough, in depth analysis and would prevent against data overload (Miles & Huberman, 1994). The month leading up to Election Day (between October 6, 2012 and November 6, 2012) was selected because past research suggests that citizens demonstrate increased attention and engagement in politics immediately prior to elections (Tolber & Mcneal, 2003). Additionally, during this timeframe the particular congressional race being examined in the present study received heavy coverage in the media as voter polls were released and candidates pursued aggressive advertising campaigns (Canham, 2012a, 2012b, 2012c; 2012d; Romboy, 49 2012a, 2012c, 2012d). Consequentially, this time frame constituted a period represented by interesting and passionate political discussion, especially in comparison to earlier points during the campaign and the weeks and months following Election Day. The digital nature of social network sites enabled the collection of all content- text, photos, videos, and hyperlinks-from the specified posts and comments by copying and pasting digital text from Facebook to an electronic document using Microsoft Word software. All data were collected from both Pages at 11:59PM on November 6, 2012 and saved in electronic computer files. Since Facebook's timeline design existed as a digital archive of activity it was possible to access previously published content from previous days and weeks. Social network site data were extracted from a publicly accessible space and therefore it was not necessary to conceal Facebook usernames during data collection and analysis procedures. Consistent with ethical guidelines for conducting Internet research, informed consent of participants is not required if the online phenomena can be accessed by anyone with an open Internet connection (Rosenberg, 2010). Although individuals were required to create a Facebook account if they wanted to publish content on Matheson's or Love's Facebook Pages, access to view content was open to anyone with an Internet connection. According to Rosenberg, "public discourse must always be open for scholarly analysis and critique, and, in lack of restricted entrance, there is no need for consent or even anonymizing" (2010, p. 24). Findings presented in subsequent chapters reflect the precise names that were attached to quotes extracted from social network sites at the time the data were collected. It should be noted that Facebook allowed users to 50 create any username and therefore users' display names were not necessarily a reflection of their true identity. The digital nature of this environment allowed the researcher to enter and exit the research site without leaving traces of observation or data collection activity. Therefore this study had no effect on the ways that Facebook users participated on the site. Further, it is important to state that the researcher refrained from publishing original content on either of the candidate's Facebook Pages in an effort to avoid manipulation of the deliberative experience being studied. Interviews. Data were also collected from interviews with campaign staff involved with social network site communication for Matheson's and Love's campaigns. Conducting in depth interviews with individuals who could describe their experience of living through the phenomenon revealed information concerning the relationship between the campaign and the deliberative activity on their social network sites. Knowledge of this relationship was crucial to gaining a deep and thorough understanding of the phenomenon in question. Because phenomenology seeks to draw a common understanding from individual life experiences, the researcher carefully selected as many interviewees as could be identified who had experienced the phenomenon in this case study (Creswell, 2007). Individuals were selected using a snowball sampling technique that began with sending an email to each campaign using the primary contact information listed on their campaign websites. The social media directors of both campaigns responded to the researcher's emails and were helpful in recommending other members of the campaign staff that were directly involved with Facebook efforts, or could offer a perspective regarding its use. In 51 an effort to collect information from as many people as possible the researcher requested interviews from 8 individuals. Due to the busy work schedules of the campaign staff, 6 of these 8 interviews were completed. Interviewees included the social media directors and the communication directors from both campaigns, as well as one candidate and one communication assistant who was involved with updating content on social network sites. Excluding the one candidate, all interviews were conducted in person at the office headquarters of each campaign. Both offices were located within the Salt Lake City area. Interviews were conducted during the week prior to Election Day from October 30, 2012 to November 1, 2012. In comparison to an earlier date, it was important to conduct interviews close to Election Day because by this time the interviewees had developed a greater understanding of Facebook use within the context of the campaign. Due to the candidates' busy preelection schedules, these interviews had to be completed following Election Day on February 6, 2013. Unlike the other interviews that were conducted in person, the one candidate reached for this study was interviewed over the phone. Prior to beginning each interview, participants were provided with an informed consent document that all interviewees reviewed and signed. This consent form ensured ethical research and was consistent with principles designated in the Belmont Report that state that anyone recruited for a study should, 1) participate on a voluntary basis, 2) be able to understand what the study will demand of him or her, 3) be able to understand the potential risks and benefits of participation, and 4) have the legal capacity to give consent (U.S. Department of Health & Human Services, 2012). The University of Utah Institutional Review Board (IRB) approved all research involving human subjects in this study. 52 All interviews were moderately structured using an interview guide. This semistructured guide allowed for flexibility to ask questions in different ways to provide clarity, as well as to allow interviewees to develop topics and raise unknown issues concerning their experiences with social network site use that they felt were important. Development of interview questions was motivated by two key questions proposed by Moustakas (1994) as being important to understanding the lived experience of participants. Key questions were: 1) What have you experienced in terms of the phenomenon? and 2) What contexts or situations have typically influenced or affected your experiences of the phenomenon? From Moustakas' broad questions several subquestions were created that were both specific to the phenomena under investigation and could help answer the research questions posed in this study. Questions were framed and directed toward unearthing participants' experiences, feelings, beliefs, and convictions about the phenomenon in question. Topics addressed in the interviews included: objectives for Facebook use, the role of Facebook in relation to other communication tools, monitoring of Facebook content, communicating with other Facebook users, the use of Facebook content for developing campaign strategy or policy decisions, and opinions regarding deliberative discourse on their Facebook Page. A final clearinghouse question was posed at the end of each interview that asked, "Is there anything else you want to tell me about Facebook or social network use in the campaign?" Topics that emerged from this last question varied considerably as it gave the interviewees an important opportunity to share some aspect of their personal experience that they felt was significant. 53 All interviews were recorded using two digital audio recording devices to ensure that they were successfully saved for subsequent analysis. Rather than relying solely on field notes, it was important to employ technology in order to capture an objective documentation of the interview discourse and to facilitate the storage, organization, and retrieval of data (Lindlof & Taylor, 2011). As explained by Lindlof and Taylor, use of audio recording allows the interviewer to engage more fully in the conversation. At the completion of each interview session digital recordings were transferred to a password protected computer and saved in electronic digital files that could be accessed for analysis at a later date. Because researchers may easily become absorbed in the data collection process and can fail to reflect on what is happening, the qualitative practice of "memoing" was also employed following each interview session (Miles & Huberman, 1994). Following Miles and Huberman's (1994) guidelines for memoing, this procedure entailed writing field notes that recorded what was heard, seen, experienced, and other thoughts that developed during the course of collection and reflecting on the interview process. These memos were dated and stored in electronic files so that they could be used later during data analysis. While individuals that were interviewed represented a range of roles and identities within the campaigns, the names and titles of interviewees were rendered irrelevant to the researcher during procedural and writing phases of this research. This allowed the researcher to gain a better understanding of the collective experience-as opposed to individual experiences-that was sought in this phenomenological case study. This decision also preserved the confidentiality of the participants who explicitly wished to 54 remain anonymous. Lastly it should be noted that all digital recordings of interviews were destroyed upon completion of this research. Data Analysis Phenomenological analysis involves a reduction process wherein the researcher takes steps to reduce the meanings of the experience to their essential structure (Creswell, 2007; Moustakas, 1994). Description is key to the phenomenologist's quest to search for all possible themes and meanings that can be conceived from the lived experience. The analysis procedure for this study was derived from the phenomenology method outlined by Moustakas (1994) and included the following steps: bracketing, horizonalization, clustering meanings, producing textual and structural descriptions, and finally producing an essential invariant structure. A brief description of each step will now be provided. Bracketing. Bracketing, or "epoche" is a process that requires the researcher to set aside their own experiences related to the phenomenon as much as possible in order to approach data with a fresh perspective. Moustakas (1994) suggested that making an explicit attempt to separate from subjective beliefs and assumptions allows the researcher to be a better receptacle of others experiences. In addition to completing this step at the beginning of the analytical process, the researcher also documented personal biases and subjectivities before each of the subsequent steps to prevent past experiences from intervening into attempts to understand others' lived experiences. Horizonalization. In this second step the researcher reads through his or her data and highlights "significant statements," sentences, or quotes that provide an understanding of how individuals experienced the phenomenon (Moustakas, 1994). Each statement, or horizon of experience, has equal worth and statements are nonrepetitive and 55 nonoverlapping. For organizational purposes computer software was employed in this step to create a spreadsheet that clearly charted the distinct statements that were extracted from the data (Creswell, 2007). More specifically, the use of a spreadsheet enabled the researcher to create a column to record significant statements, and to create as many subsequent columns as necessary to record notes and codes related to a given statement. For example, when analyzing citizens' comments published on social network sites it was useful to create a column that contained significant statements, as well as an additional column that contained the exact text and multimedia from the post that stimulated the significant statement. This organizational scheme enabled the researcher to quickly recall the context in which a significant statement was made, and also allowed for easy reference to the day and post where the statement was originally published. Because social network site data are originally produced in a digital format, it was simple to copy and paste this content into a spreadsheet. However, to analyze interview data it was necessary to first process digital audio recordings through a transcription procedure. Transcribing entailed listening to recordings of each interview and transcribing key words, phrases and statements. Participants' exact quotes were copied as much as necessary in an effort to allow the voices of the participants to speak, which was conducive to generating a more accurate understanding of individual experiences. From these transcriptions, significant statements were extracted and added to the analysis spreadsheet. Clustering meanings. In this third step the researcher clusters significant statements to develop themes. By interrogating the meaning of the various clusters, central themes were determined "which expresses the essence of these clusters" (Hycner, 56 1999, p. 153). In this stage it was necessary to navigate back and forth through the list of significant statements created in the previous step, as well as original data recordings and memos so that themes were accurately formulated to represent lived experiences of democratic deliberation on campaign social network sites. Themes constructed from clustering significant statements were then further condensed to create categories of major contextual issues that were identified during observation and analysis of the phenomenon. Textual and structural description. In this step composite summaries were written to describe the textual and structural properties of the major contextual issues that were strategically categorized in the previous step. These summaries explained what happened when the phenomenon was experienced, and also contained explanation of the context or setting of the experience to reveal how the phenomenon was experienced. As a result of following this phenomenological method, textual and structural descriptions accurately reflected the context or "horizon" from which the themes emerged (Hycner, 1999; Moustakas 1994). Textual and structural descriptions are presented in Chapter 4. For organizational and clarity purposes, these summaries were organized around the research questions posed in Chapter 1. Essential invariant structure. This final step involved drawing on the textual and structural descriptions produced in the previous step to write a composite description that presented the essence of the phenomenon. More specifically, the aspects of the experiences that are common to all participants are invariant structures and revealed the essential nature of the phenomenon. This summary is contained in Chapter 5. 57 Figure 1. Screen shot of Matheson's campaign Facebook Page captured from http://www.facebook.com/MathesonUT on October 22, 2012. This figure illustrates the structure and aesthetic elements that appeared on Matheson's Facebook Page. 58 Figure 2. Screen shot of Love's campaign Facebook Page captured from http://www.facebook.com/mialove on November 4, 2012. This figure illustrates the structure and aesthetic elements that appeared on Love's Facebook Page. 59 Figure 3. Example of Facebook "post" and "comments" captured from http://www.facebookmialove on November 4, 2012. This figure illustrates what a post and comments look like on Facebook Pages. CHAPTER 4 RESULTS The results presented in this chapter reveal characteristics of the nature of democratic deliberation conducted on two electoral candidates social network sites. As explained in Chapter 3, the phenomenological data analysis method that was employed in this study produced findings that describe the textual and structural properties of the major contextual issues surrounding the democratic deliberation examined in this case study. In this chapter, composite summaries that were produced from the analytical clustering procedure prescribed in Chapter 3 are organized to respond to the research questions posed in Chapter 1. Following a concise explanation of the meaning and significance of each category, relevant examples and quotes that were extracted from data will be presented in an effort to illustrate the function of the category in the context of an actual deliberative scenario. Strategies Used by Citizens to Engage in Democratic Deliberation The first research question asked: What strategies do citizens use to engage in democratic deliberation on electoral candidates social network sites? Findings responding to this question were derived from scrutiny of the social network site data collected in this case study. To review, these data included all multimedia text and images published during the month preceding Election Day on the campaign Facebook Pages of Utah 61 congressional candidates Jim Matheson and Mia Love. As expected, democratic deliberative discourse that was consistent with the researcher's conceptualization posed in Chapter 1 (democratic deliberation was defined as public citizen discourse, in the form of formal and informal exchanges, that address issues of public concern, where disagreement exists and a collective decision is needed) was observed throughout the data collected as citizens used the public Facebook Pages of the two electoral candidates to present their political perspectives and respond to other's opinions. Additionally, the researcher's anticipation of lively debate and disagreement was significantly represented in exchanges between citizens featured on both candidates' sites. A diversity of opinions was reflected in deliberation as citizens articulated varying levels of support for candidates, policy issues, political ideologies, and democratic processes. Before proceeding to a description of findings it must be acknowledged that in addition to identifying deliberative communication on the social network sites, the researcher also observed other types of communication that did not fit the conceptualization of democratic deliberation used in this study. For example, brief expressions of support or opposition to candidates were observed throughout the data collected. To demonstrate, comments from a post published on Mia Love's campaign Page on November 6, 2012 included the following expressions of support that were not classified by the researcher as contributing to deliberative discourse: Rich Edmiston wrote, "Good Luck Mia"; Damon Troy Allen wrote, "God Speed today Mia…!!!"; and Lily Gonzalez wrote, "You go girl!!!" (Mia Love, 2012d). Likewise, this post also featured the following expressions of opposition to Love: Arnett Gayle wrote, "Nice try but, yu [sic] gonna loose"; Stephen Murdock wrote, "barf"; and Fred Ketterer wrote, "No 62 thanks not my flavor" (Mia Love, 2012d). While these types of comments may contain value for other studies concerning political communication on social network sites, they were not evaluated in the present study as the research questions and method used in this investigation sought to concentrate on understanding the nature of democratic deliberation. Analysis procedures produced seven categories of findings that represent the strategies used by citizens to engage in democratic deliberation on the two electoral candidates social network sites. The researcher will now elaborate on each of these categories and provide examples to illustrate analytical descriptions. In the subsequent presentation of findings, direct quotations were extracted from the Facebook Pages of both candidates in the form of individual statements and discursive exchanges. All quotes are cited using the Facebook usernames that were associated with each comment as they appeared on the Page. The post that the content was featured within will also be referenced. When presenting discursive exchanges involving more than one citizen, the time that the comment was published will also be cited in an effort to provide some context for the deliberation. Citizens Use Hyperlinks to Support Opinions Citizens employed the use of hyperlinks to support their opinions when participating in deliberation on candidates' Facebook Pages. To understand the significance of this digital strategy it is first necessary to explain the operation and purpose of a hyperlink. When featured on a website, a hyperlink, or a link, is an element such as a word, phrase, or image that links to another place on the same site or to an entirely different site. Hyperlinks facilitate quick and efficient Internet browsing because 63 clicking on a link redirects a user to a different web location. In the present case study, deliberators embedded hyperlinks in the body of their comments to offer additional information that pertained to, and supported their arguments. The use of hyperlinks in deliberative discourse is unique to digital communication platforms and emerged in this study as a strategy for enhancing written opinions. Three subcategories of hyperlinking were identified that included: linking to different Facebook Pages, linking to stories on news outlets' websites, and linking to external websites. Hyperlinking to different Facebook Pages. Citizens published comments that included hyperlinks to other Facebook Pages operated by individual persons or groups within the Facebook social network site. On Facebook, hyperlinked text was differentiated as blue, in comparison to plain nonhyperlinked text that was black. Clicking on hyperlinked text redirected the user to a different Page in the social network site that may have included a range of content such as photos and other information that contributed to supporting the opinion being conveyed by the deliberator. In the data analyzed deliberators linked to Facebook Pages operated by persons or groups that may or may not have been participating in the deliberative discourse occurring. An example of linking to a Page belonging to a group that was not involved in transpiring deliberative discourse was identified in a comment published on Jim Matheson's campaign Page by Facebook user Donald Bush. Bush's comment included the following hyperlink to a story about Love that was published on the Facebook Page of the Republican Security Council: http://www.facebook.com/photo.php?fbid=327637117335466&set=a.265394146893097. 64 53820.101173139981866&type=1&theate (Matheson for Congress, 2012a)1. The linked Republican Security Council Page reported results from three voter polls that showed Love defeating Matheson in the electoral race. The linked Page also featured a photo of Love as well as a positive statement about her political character. In this example, embedding a hyperlink to a different Facebook Page functioned to provide additional information that supported Bush's opinion regarding Love. Citizens also provided hyperlinks to the Facebook Pages of other individuals or groups involved in transpiring deliberation in an effort to identify or respond to a specific comment. For example, in deliberation involving multiple Facebook users regarding the policy positions of the presidential candidates, Facebook user Jennifer Jensen Davis started her comment by including a hyperlink to Elizabeth Eyes Delgado Clark who was also participating in the deliberation (Matheson For Congress, 2012b). Davis' hyperlink functioned to let the other Facebook users involved in the conversation know that she was responding to Clark specifically. Including a link to Clark's Facebook Page also enabled readers to easily view Clark's profile information and learn more about her by simply clicking on the hyperlink. As demonstrated in these examples, hyperlinking to other Pages within the Facebook social network site was a strategy used by citizens to reference information and draw attention to specific persons' or groups' Facebook profiles. Hyperlinking within the Facebook site was significant because it allowed readers to be redirected to profile Pages 1 Guidelines for referencing content from social network sites such as Facebook were not included in the American Psychological Association's publication style manual at the time when this study was conducted. Therefore, the style used for citing Facebook content was derived from the method used to cite a website. Material from Facebook is referenced in this study according to the post where the content was originally published. 65 that featured background information about users. The ability to access profiles of other deliberators is a characteristic of social network sites that is distinct from other deliberative venues. Hyperlinking to news stories. Citizens embedded hyperlinks in their comments that linked to news stories published in online newspapers such as The New York Times, Politico, and Deseret News. For example, Facebook user Peggy Wilson used this strategy to explain why Love should not support presidential candidate Mitt Romney (Mia Love, 2012a). To elaborate, Wilson embedded a hyperlink to a story on The New York Times website that discussed Romney's position to eliminate emergency management disaster coordination. Inclusion of this hyperlink allowed Wilson to provide evidence for her opinion without having to write a long comment that summarized the contents of the article. Hyperlinking to the website of a trusted news outlet such as The New York Times also functioned to enhance the credibility of Wilson's opinion. Hyperlinking to external websites. Citizens included hyperlinks in the text of their comments that redirected the users to a variety of external websites such as personal blogs, interest groups, government sites, Wikipedia, and campaign websites of electoral candidates. For example, in deliberation concerning the political partisanship of Matheson supporter Bill Applegarth, Facebook user Riverton Utah wrote that Applegarth was a "RINO" (Matheson for Congress, 2012c). To elaborate on what is meant by this term, Riverton Utah presented a hyperlink to the "Republican in Name Only (RINO)" Wikipedia Page, which displayed additional information regarding the origin and usage of this term. This use of a hyperlink by Riverton Utah simultaneously helped to explain 66 the language expressed in his or her comment and provide support for the opinion expressed. Another example of this subcategory of hyperlinking was observed in deliberation about funding special education that was engaged on Jim Matheson's campaign Page. Facebook user Cathie Chansamone Costanzo wrote the following: I am mom to a severely autistic child. I fight for funding for programs and being a child of a vet, and former federal govt. [sic] worker, every child in the state gets extra federal funding for programs to help keep local property taxes down. We also, thanks to grants and student aid, have a very highly educated workforce and high school graduation and univ [sic] attendance rates in the USA. Source, www.ed.gov the agency that has ensured this in our state. [sic] and [sic] nces.ed.gov. they [sic] are created to ensure the equality and enforce the anti discrimination laws, that without, she [Mia Love] would not have been able to attend college let alone run for office as a minority. I am a minority and those laws have helped many improve quality of life and for the disabled before [sic]. (Matheson for Congress, 2012d) In this quote Costanzo supplied hyperlinks to two external government websites, www.ed.gov and nces.ed.gov, in an effort to cite where she retrieved the facts to support her opinion. Instead of simply stating the source of information or presenting facts without citing a source, Costanzo used a hyperlink to defend and add credibility to her claims. This allowed other Facebook users to check the validity of Costanzo's statement and also access further statistics and information related to the topic. Similar to the function of the other subcategories in this category, hyperlinking to external websites was a strategy used by citizens to enhance their arguments and advance their opinions in deliberative discourse. Citizens Use Content in Campaign Posts to Guide Deliberation Citizens drew on the content featured in Facebook posts (that was published by the campaigns) to engage in democratic deliberation. Specific topics, policy issues, 67 people, and places that were depicted in posts appeared to stimulate subsequent comments that reflected deliberation related to particular topics addressed in posts. Analysis revealed that citizens responded to questions and statements contained in posts, and also used this content to develop their opinions about candidates. To further explain the significance of this category, findings were divided into two subcategories based on the type of content contained in posts. News articles guide deliberation. Both Matheson's and Love's campaign Facebook Pages exhibited posts that featured hyperlinks to online news articles written about the 4th District election, or one of the candidates specifically. Inclusion of a hyperlink in a Facebook post allowed users to easily click on the link and be redirected to the news outlet's website where the article was originally published. Hyperlinked news outlets observed in this case study included online periodicals such as Deseret News, The Salt Lake Tribune, and U.S. News & World Report, as well as television and radio stations such as KUER, KSL, CNN, and Utah's Fox 13 News. Statements, quotes, and photos contained in the hyperlinked news articles existed as fodder for deliberation. For example, on October 24, 2012 Jim Matheson's campaign published a post that featured a link to a news article written in The Salt Lake Tribune and a quote extracted from the article that said: ‘To suggest that somehow Utah can make do without this, go talk to the people who run these special ed programs in our schools. …It's not feasible, it's not realistic and it's an uninformed position on the part of my opponent,' Matheson said. (Matheson for Congress, 2012d) In the subsequent comments published under this post, several different Facebook users deliberated Love's and Matheson's positions on funding for special education and the issue of special education in general. Drawing on content from The Salt Lake Tribune 68 article, Facebook users presented their opinions and asked questions related to the issue. For example, Facebook user Ray Rizer questioned the position of Love writing, "Am I misinterpreting her walk-back? Is she saying the federal funds should go directly to the states for special ed?" (Matheson for Congress, 2012d). A different Facebook user Ryan Lufkin quoted the article to support his opinion writing, "Earlier this year, Love put out a fairly specific proposal to cut federal spending that included eliminating all federal education subsidies.' Sounds like Mia Love is the one who should be ashamed..." (Matheson for Congress, 2012d). A third Facebook user Beverly Nelson Martinez offered her opinion regarding the issue stating: What would our kids do without special education? flounder! and what would regular ed. [sic] teachers do with all of those kids with severe disabilities in their classrooms? whole classrooms would fl