Political Blogs: Transmission Belts, Soapboxes, Mobilizers, or Conversation Starters?

Journal of Information Technology & Politics ISSN: 1933-1681 (Print) 1933-169X (Online) Journal homepage: http://www.tandfonline.com/loi/witp20 Political Blogs: Transmission Belts, Soapboxes, Mobilizers, or Conversation Starters? Kevin Wallsten To cite this article: Kevin Wallsten (2008) Political Blogs: Transmission Belts, Soapboxes, Mobilizers, or Conversation Starters?, Journal of Information Technology & Politics, 4:3, 19-40, DOI: 10.1080/19331680801915033 To link to this article: https://doi.org/10.1080/19331680801915033 Published online: 11 Oct 2008. Submit your article to this journal Article views: 1040 View related articles Citing articles: 33 View citing articles Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalinformation?journalcode=witp20

WITPolitical Blogs: Transmission Belts, Soapboxes, Mobilizers, or Conversation Starters? Wallsten Kevin Wallsten ABSTRACT. This paper makes an initial attempt to situate political blogging alongside other forms of political participation by asking the question: how do political bloggers actually use their blogs? More specifically, this paper relies on a detailed content analysis of 5,000 less popular and 5,000 A-list political blog posts over the course of the 2004 campaign in order to determine whether political bloggers use their blogs primarily as soapboxes, transmission belts, mobilizers, or conversation starters. The results presented here suggest that although political blogs are used to make opinion statements far more often than they are used to mobilize political action, to request feedback from readers, or to pass along information produced by others, blog use changes significantly in response to key political events. To be more precise, less popular political bloggers were significantly more likely to mobilize political action on Election Day, and all bloggers regardless of popularity showed a greater propensity to seek feedback from their readers on the days of the presidential debates and in the weeks immediately following the election. Political blogging, in short, is a complex form of political participation that blends hypertext links, opinionated commentary, calls to political action, and requests for feedback in different ways at different moments in time. KEYWORDS. Blogs, political participation, mobilization, 2004 election INTRODUCTION THE BLOGGING EXPLOSION Blogging the act of creating and maintaining an online diary where information is chronologically posted, updated frequently, and presented in reverse chronological order (Blood, 2000) has become an incredibly popular activity in recent years. 1 In 1999, the total number of blogs was estimated to be around 50 (Drezner & Farrell, 2004a). In 2002, a Newsweek article estimated the total number of blogs at 500,000, with a new blog starting every 40 seconds (Levy, 2002). Despite how extravagant these early estimates seemed at the time, they turned out to be excessively conservative. In 2003, a survey by the Perseus Development Corporation, for example, found that there were 4.12 million blogs (Henning, 2003). A study by the Pew Internet and American Life Project found that 7% of the 120 million Internet users in the United States had created a blog by November 2004, and by January 2005 that number had increased to 10%. Similarly, the number of blogs tracked by Technorati (http://www.technorati.com), a Web site devoted to measuring activity in the blogosphere, has doubled every five months Kevin Wallsten is a Ph.D. candidate in the department of political science at the University of California, Berkeley and a fellow of the Miller Center of Public Affairs. The author would like to thank Laura Stoker, Tatishe Nteta, Jill Greenlee, Rachel Van Sickle-Ward, Jack Citrin, Antoinette Pole, and the anonymous reviewers for comments on previous drafts of this article. Address correspondence to: Kevin Wallsten, Department of Political Science, University of California, Berkeley (E-mail: wallsten@berkeley.edu). Journal of Information Technology & Politics, Vol. 4(3) 2007 Available online at http://jitp.haworthpress.com 2007 by The Haworth Press. All rights reserved. doi:10.1080/19331680801915033 19

20 JOURNAL OF INFORMATION TECHNOLOGY & POLITICS since 2003. At the time of this writing (March 2007), Technorati is currently tracking over 70 million blogs. Following this overall growth in blogging, it appears that the number of people engaging in explicitly political blogging has also increased in recent years. As Figure 1 shows, the number of political blogs listed on the Eatonweb Portal (http://www.portal.eatonweb.com), 2 a popular blog directory, has increased dramatically since 2001. Interestingly, political blogging seems to have grown faster than other kinds of blogging. Indeed, as Figure 1 also shows, the number of political blogs has increased at a greater rate than the number of art, technology, business, music, and sports blogs. According to a recent survey by the Pew Internet and American Life Project, there are currently 1.4 million blogs that contain purely political information (Lenhart & Fox, 2006). 3 This paper seeks to shed light on a small part of the political blogging phenomenon by asking the question: how do political bloggers use their blogs? More specifically, this paper relies on a detailed content analysis of 5,000 less popular and 5,000 A-list political blog posts over the course of the 2004 campaign in order to determine whether political bloggers use their blogs primarily as soapboxes, transmission belts, mobilizers, or conversation starters. The results presented here suggest that although political blogs are used to make opinion statements far more often than they are used to mobilize political action, to request feedback from readers, or to pass along information produced by others, blog FIGURE 1. Blog Growth by Type, February 2001 to September 2006. 8000 7000 6000 Number of Blogs 5000 4000 3000 2000 1000 0 Feb-01 Jun-01 Oct-01 Feb-02 Jun-02 Art Entertainment Politics Sports Oct-02 Feb-03 Jun-03 Oct-03 Feb-04 Jun-04 Oct-04 Feb-05 Jun-05 Oct-05 Feb-06 Jun-06 Technology/Computers Finance/Business Music Source: Eatonweb Portal

Wallsten 21 use changes significantly in response to key political events. To be more precise, less popular political bloggers were significantly more likely to mobilize political action on Election Day, and all bloggers regardless of popularity showed a greater propensity to seek feedback from their readers on the days of the presidential debates and in the weeks immediately following the election. Political blogging, in short, is a complex form of political participation that blends hypertext links, opinionated commentary, calls to political action, and requests for feedback in different ways at different moments in time. LITERATURE REVIEW Unsurprisingly, the rapidly expanding number of blogs has been accompanied by a surge in the amount of research on political blogging. While the vast majority of scholarly attention has focused on the linkage patterns between political blogs (Ackland, 2005; Adamic & Glance, 2005; Drezner & Farrell, 2004a; Hargittai, Gallo, & Kaine, 2007), and on the impact that blogs have on mainstream media coverage (Drezner & Farrell, 2004b; Hewitt, 2005; Roth, 2004; Schiffer, 2005; Smolkin, 2004; Wallsten, 2006), election campaigns (Bloom & Kerbel, 2007; Trippi, 2004; Williams, Trammell, Postelnicu, Landerville, & Martin, 2005), legislative politics (Bloom, 2003; Sroka, 2006; Wright, 2003), and international affairs (Drezner & Farrell, 2004b; McKinnon, 2007; Zuckerman, 2007), there is a growing literature that seeks to empirically assess how political bloggers use their blogs. Do political bloggers, for example, devote most of their blog posts to mobilizing political action, or do they just pass along information produced by others? Similarly, are political bloggers frequently trying to engage in dialogue with their audiences, or do they spend most of their time spelling out political positions in an attempt to persuade anyone who might be reading? Using data derived from interviews with non-political bloggers, a number of studies have attempted to address questions such as these by exploring the reasons that people blog. Nardi, Schiano, Gumbrecht, & Swartz (2004), for example, found that bloggers blog in order to document their personal lives, to express their feelings, to interact with like-minded people, and to provide readers with commentary and opinions. Similarly, a recent survey of a random sample of bloggers found that expressing yourself creatively, networking, sharing knowledge, motivating others to take action, and influencing the way other people think were all major reasons why individuals create and maintain blogs (Lenhart & Fox, 2006). Mimicking these results amongst a much different group of bloggers, Pole (2006) found that African American bloggers use their blogs to encourage donations to philanthropic causes, to advocate for political causes, and to inform their readers about errors or omissions in the mainstream media. More germane to the purposes of this paper, a number of studies have analyzed how explicitly political bloggers use their blogs. In their research on popular political bloggers, for example, McKenna and Pole (2004) found that popular political bloggers blog because it provides them with an opportunity to add new voices to the political debate, to increase political activism, to engage in dialogue with other citizens, and to expose readers to new sources of information. In a similar study of less popular political bloggers, McKenna and Pole (in press) found that less popular political bloggers use their blogs to inform readers, to advocate for causes, and to attempt to mobilize political action. Looking at a sample of bloggers who focus their blogging on one specific issue, McKenna (2007) found that so-called policy bloggers use their blogs to filter information, to provide expertise, to form networks, to attract attention, to frame arguments, and to exploit windows of opportunity. Taken together, these studies suggest that political bloggers use their blogs to express their political beliefs, to interact with likeminded people, to inform their readers, and to influence the political world around them. As a result of the fact that they rely solely on crosssectional interview data; however, these studies provide only a snapshot of how political bloggers use their blogs and not specific measures of how often political bloggers use their blogs for different activities at different points in

22 JOURNAL OF INFORMATION TECHNOLOGY & POLITICS time. To be more precise, the reliance of these studies on cross-sectional interview data makes it impossible to determine what percentage of blogging activity is devoted to expressing opinions, interacting with readers, passing along information and attempting to influence the political world and, more importantly, whether political bloggers turn their political blogs to different purposes in response to changing political events. In addition to fundamentally limiting our understanding of the nature of the blogging phenomenon, the lack of empirical evidence regarding blog usage is unfortunate because it deprives researchers of clues about where to look when attempting to assess the political impact of political blogs. If, for example, political bloggers devote most of their blog posts to raising money for candidates, organizing boycotts against media institutions, and recruiting volunteers for political campaigns, researchers should focus on determining whether blogs increase political participation among readers. If, however, attempts to mobilize political action are rare and posts on political blogs are usually reserved for the blogger s opinionated commentary, researchers should concentrate their attention on measuring the extent to which blog readers be they members of the mass public, journalists, or politicians are persuaded by the statements they read on political blogs. In short, determining what political bloggers choose to do with their blogs at different points in time is a necessary step to creating and testing theories about the consequences of political blogging. In order to better assess the ways that political bloggers actually use their blogs, this paper relies on a longitudinal content analysis of political blogs. More specifically, I developed a coding scheme for classifying blog use and applied this scheme to a sample of the posts found on political blogs from July 1, 2004 to November 30, 2004. Although limiting the analysis to such a short period of time necessarily constrains the kinds of conclusions that can be drawn about how political bloggers use their blogs, 4 the selection of this time period is strategic and comes with two main benefits. First, studying this period provides a baseline for political blog use during the first Presidential Election in the United States in which blogging played an important role (Adamic & Glance, 2005, p. 1). Second, and more importantly, studying this period allows for an assessment of the extent to which political blog use changes in response to different political conditions because it includes a number of weeks before the presidential campaign attracted much attention (the days prior to the start of the Democratic National Convention on July 26, 2004), a number of weeks during the height of the campaign (the days between the Democratic National Convention and the election on November 2, 2004), and a number of weeks following the election (the days after November 2, 2004). This paper, therefore, builds on previous research into political blogging by using content analysis to determine the precise percentage of blogging activity that is devoted to various kinds of activities and whether this percentage changes over time. TRANSMISSION BELTS, SOAPBOXES, MOBILIZERS, AND CONVERSATION STARTERS A defining feature of blogs and, no doubt, a key part of their popularity is that they are subject to almost no external editing and, therefore, provide the blogger with complete control over when and what to post. 5 As a result, cataloguing all of the things that bloggers use their blogs for is difficult if not impossible. Generally speaking, however, there are four ways that a blogger committed to political blogging might use his or her blog. First, political blogs can be used as link filters or transmission belts that simply provide links to Web sites or quote sources with no commentary from the blogger. Using a political blog as a link filter or transmission belt is designed to provide the reader with an efficient way to separate out the interesting political stories and links from the vast number of uninteresting stories and links without explicitly injecting the blogger s own opinion into the reader s search for information. The emphasis for bloggers who use their blogs in

Wallsten 23 this way is in knowing what is going on in the political world outside of the site and in trying to help readers find it. Second, and standing in stark contrast to the political blog as transmission belt, political blogs can be used as technological soapboxes from which bloggers spout their opinions into the vast expanse of cyberspace. Using a political blog as a soapbox is designed to provide the reader with a record of the thoughts, observations, experiences, and opinions of the blogger. As a result of the fact that they focus on the blogger s internal reaction to political developments, blogs of this kind may resemble a political diary or a political confessional. Third, political blogs can be used as mobilizers that try to encourage readers to take political action. Using a political blog as a mobilizer is designed to lower the costs of political participation for the blog s readers, thus increasing the likelihood of participation by providing an easy source of information about political events and opportunities for political action. It should be clear that using a political blog as a mobilizer is not mutually exclusive with using a political blog as a transmission belt or using a political blog as a soapbox. Indeed, bloggers can attempt to mobilize political action by simply posting information from other sources without further comment or they can spell out detailed arguments in favor of a particular cause before suggesting that readers take political action. Finally, political blogs can be used as conversation starters that try to elicit feedback from readers. There are a number of ways that bloggers can use their blogs as conversation starters. In addition to inviting feedback by providing an e-mail address and a comments section, bloggers can encourage dialogue with readers by asking direct questions to their readers in the body of their posts. Similar to using political blogs as mobilizers, using political blogs as conversation starters is not inconsistent with using political blogs as soapboxes. Indeed, oftentimes requests for reader feedback will be preceded by a detailed discussion of the blogger s views on a certain political issue. Moreover, using political blogs as conversation starters can fit with using political blogs as mobilizers. Bloggers may, for example, encourage feedback from readers at the same time that they encourage readers to take political action in other ways. In short, political blogs may simultaneously act as soapboxes, mobilizers, and conversation starters. THE SAMPLE In order to explore whether political blogs act as soapboxes, mobilizers, transmission belts, or conversation starters for a wide variety of political bloggers, a sample of less popular political blogs and a sample of popular, A-list blogs was collected. Gathering a sample of less popular political blogs presents a more difficult problem than gathering a sample of A-list political blogs. Indeed, whereas the number of popular political blogs is relatively small and there are a few, well-known sites that are devoted to tracking who is on and who is off the A-list, the number of less popular political blogs is literally countless and there is no single Web site that claims to track all less popular blogs. Fortunately, however, many bloggers choose to list their blogs on one of the many blog directories that have sprung up around the Internet and, more importantly, these directories allow bloggers to categorize their blogs based on the subjects the blogger thinks their blog discusses most. Since these directories allow for searches based on these subject keywords, it can be relatively easy to locate less popular blogs that are political. Despite the fact that these directories include only those blogs that have been submitted for inclusion by their authors and, as a result, cannot produce anything like a complete list of political blogs from which to sample, I generated a population list of 10,732 unique political blogs from 12 6 of the most well known blog directories. 7 From this list, I made a random selection of 250 political blogs. 8 Because this paper is concerned only with American political blogs, I determined if the blog written by an author in the United States. 9 Of the 250 less popular political blogs sampled, 12 were not actually blogs, 33 had addresses that no longer worked, 58 blogs were not written by authors in

24 JOURNAL OF INFORMATION TECHNOLOGY & POLITICS the United States and 147 were written by authors in the United States. 10 Although collecting a sample of A-list blogs presents far fewer methodological problems than collecting a sample of less popular political blogs, constructing a population list of A- list blogs from which to sample is complicated by the fact that there is no agreed upon system for ranking the popularity of a blog. As a result, a number of different sites that rely on a number of different criteria have emerged to rank the popularity of blogs. Blogstreet (http:// www.blogstreet.com), The Truth Laid Bear Ecosystem (http://truthlaidbear.com/ecosystem.php), and Technorati, for example, compile their rankings based on how many times a blog is linked to another blog. By contrast, The Truth Laid Bear Ecotraffic (http://truthlaidbear.com/ ecotraffic.php) compiles its rankings based on the number of hits each blog receives. 11 In order to create a population list of A-list blogs from which to sample, I downloaded the top 100 rankings from these four sources (Blogstreet, The Truth Laid Bear Ecosystem, The Truth Laid Bear Ecotraffic, and Technorati) during the first week of February 2005. I included each blog that appeared on at least two of the four Web sites top rankings on the final population list. 12 From this population list of 84 blogs, I randomly selected 20 blogs. Because this paper is concerned only with the content of political blogs and, more specifically, the content of American political blogs, I checked each of the 20 sampled blogs to ensure that they discussed political issues and that they were written by authors in the United States. 13 I did not include blogs that did not meet these two criteria in the analysis that follows. Overall, 3 blogs did not discuss political issues, 1 blog was written by an author outside the United States, and 16 blogs discussed political issues and were written by authors inside the United States. 14 The final sample of A-list blogs is included in Appendix A. Before discussing the ways that the political blogs in these samples were used during the 2004 campaign, it is useful to say a few words about the blogs themselves. First, the sample of A-list blogs appears to slightly over-represent liberal blogs. Using the coding scheme presented in Appendix B, each of the 16 A-list blogs in the sample was coded for ideology. Overall, there were seven liberal blogs (44%), seven conservative blogs (44%), and two blogs with no clear ideological position (12%). In order to determine whether this sample was representative of all A-list blogs, each of the 69 A-list blogs that were written by authors in the United States and that focused on political issues was coded for ideology. Of these 69 A-list blogs, there were 25 liberal blogs (36%), 32 conservative blogs (46%), 2 independent blogs (3%), and 10 blogs with no clear ideological position (14%). 15 As a result of the sampling error produced by the relatively small size used in this study, therefore, liberal blogs are slightly overrepresented, while blogs with less clear ideological leanings and independent blogs are slightly underrepresented. 16 Second, the number of liberal blogs in the sample of less popular political blogs far exceeds the number of conservative or independent blogs. To be more precise, there were 67 liberal blogs (46%), 44 conservative blogs (30%), and 11 independent blogs (7%) in the sample of 147 less popular political blogs. 17 In addition to being relevant to the specific goals of analyzing political blog use during the 2004 campaign, the results of this ideological coding are also interesting because they provide some initial insight into the distribution of ideological commitments in the blogosphere. While further research is needed to explore whether the large number of liberal bloggers in the sample is merely a result of a tendency on the part of conservative bloggers to avoid posting their blogs on blog directories, it appears that liberals have adopted political blogging in far greater numbers than either conservatives or independents. 18 These findings are consistent with previous research that has shown Internet activists are actually more likely to be liberal than conservative or independent (Hill & Hughes, 1997). Third, there were significant differences between the level of activity on the A-list blogs in the sample and on the less popular blogs. To be more precise, A-list bloggers posted an average of 10.5 times per day while less popular bloggers posted only 1.6 times per day. Although this finding is not surprising given

Wallsten 25 that A-list blogs are popular, in part, because they are frequently updated with new content to attract readers, there was also significant variation in the level of activity within each sample. Among the A-list blogs in the sample, for example, Informed Comment (http://www. juancole.com) averaged only 2.7 posts per day while Instapundit (http://www.instapundint. com) posted an average of 20.6 posts per day. Similarly, among the less popular political bloggers, Random Thoughts (http://www. snu nes.blogspot.com) averaged 34.3 posts per day while The Whine Rack (http://www.whinerack. com) posted only 34 times during the whole period of the study. In short, political bloggers, regardless of popularity, devote vastly different amounts of time and attention to their political blogs. Fourth, although all of the blogs in the sample of A-list blogs dealt almost exclusively with political issues, the blogs contained in the sample of less popular blogs varied widely in the overall level of political discussion. Some of the blogs in the final sample of less popular blogs were exclusively political and offered almost no details on the personal life of the blogger, while other blogs doubled as personal blogs, with stories and anecdotes about the blogger s personal life filling a large portion of the blog. This variation in the amount of political discussion is the inevitable product of the fact that blogs allow authors to discuss any topic they want and that blog directories impose no restrictions on the keywords that can be used to describe one s blog. During the period of study, the 16 A-list blogs included in the sample contained 24,307 posts and the 147 less popular blogs contained 27,079 posts. Ideally, a study of how political bloggers use their blogs would proceed by coding all of these posts for whether they fit into the soapbox, mobilizer, transmission belt, or conversation starter categories described above. Given the detailed nature of the content analysis that is required to make these classifications, however, it is not possible to code every one of the 51,386 posts included on the blogs in the sample. As a result, I randomly selected 5,000 posts from the 24,307 A-list blog posts and 5,000 posts from the 27,079 posts less popular blog posts. Such large sample sizes insure that the findings presented below will be representative not only of the overall population from which the posts are drawn but also of the individual blogs from which they are drawn. MEASURING BLOG USE In order to assess how bloggers used their political blogs during the 2004 campaign, each of the 5,000 posts in the sample of A-list blogs and each of the 5,000 posts in the sample of less popular blogs was coded for whether the post merely provided a link or quoted a source without commentary from the blogger, whether the post included some statement of personal preference, whether the post asked readers to take some form of political action, and whether the post asked readers for feedback. Since providing a link without commenting and quoting a source without commenting are mutually exclusive from providing commentary and requesting feedback from readers, 19 there are nine possible classifications for each post included in the sample: (1) link or quote only; (2) commentary only; (3) mobilize only; (4) request for feedback only; (5) commentary and mobilize; (6) commentary and request for feedback; (7) commentary, mobilize, and request for feedback; (8) mobilize and request for feedback; and (9) mobilize through link or quote. 20 In order to illustrate the differences between these categories, Table 1 presents examples of posts from each of the first seven categories taken from the sample of A-list blogs. 21 A more detailed discussion of the coding scheme used to categorize each post can be found in Appendix C. To the extent that posts fall in categories 1 and 9, political blogs can be thought of primarily as link filters or transmission belts. Similarly, to the extent that posts fall into categories 2, 5, 6, and 7, political blogs can be thought of as soapboxes. To the extent that posts fall into categories 3, 5, 7, 8, and 9, political blogs are best thought of as mobilizers; and to the extent that posts fall into categories 4, 6, 7, and 8, political blogs are best thought of as conversation starters.

26 JOURNAL OF INFORMATION TECHNOLOGY & POLITICS TABLE 1. Examples of Blog Posts by Category Category Examples of blog posts (1) Link or Quote Only LT SMASH has a roundup of blog commentary on Team America. Instapundit October 19, 2004 (2) Commentary Only Another Slimy Rhodes Scholar Politician Democrat Brad Carson is running for Senate in Oklahoma against one of my all-time favorite conservatives, Tom Coburn. Carson is a Rhodes Scholar with a goodytwo-shoes image....carson is also a slimy campaigner. And a punk.... Michelle Malkin November 1, 2004 (3) Mobilize Only Hoeffel Need People at 5 th and Market If anyone can get there now to hold signs and cheer in front of KYW studios.... Eschaton November 2, 2004 (4) Feedback Only Open Thread The Debate is On Chat away.... America Blog September 30, 2004 (5) Commentary and Mobilize LGF the Vote! It s vitally important that every LGF reader get out there to your polling place and vote... Do you want to see John Kerry going hat in hand to the UN and posing for photos with Kofi Anan... Do you want to see the Arab world partying like it s 9/11?.... Go vote now! Go! Little Green Footballs November 2, 2004 (6) Commentary and Request Feedback Chris Matthews lost it yet again on Hardball last night. Anyone have a transcript? Vodka Pundit August 19, 2004 (7) Commentary, Mobilize and, Feedback Why do these companies help the Salvation Army bigots? The Salvation Army.... lobbies the government in favor of anti-gay legislation... Target is no longer letting the Salvation Army use its stores to solicit funds. The following companies are filling in the gap..... If you guys can get me the contact info for these companies, I ll post it on this site so we can urge them to stop... This holiday season DO give to a charity but don t give to a bigoted fundamentalist church. America Blog November 15, 2004 Less Popular Blogs RESULTS To what extent are political blogs actually used as transmission belts, soapboxes, mobilizers, or conversation starters? Figure 2 displays the overall percentage of posts on less popular political blog posts that fall into each of the nine categories described above. As Figure 2 shows, approximately 15% of the posts in the sample were classified as link or quote only posts, meaning the blogger merely provided a link or quoted a source without adding his or her own commentary, while the vast majority of posts (83%) were classified as commentary only, meaning the blogger stated his or her opinion about some issue but did not try to mobilize readers or elicit feedback from them. As Figure 2 also shows, the less popular political bloggers in the sample did not ask much of their readers. Indeed, requests for feedback appeared in only about 1% of posts in the sample, and attempts at mobilization occurred in only 1.9% of posts. Unsurprisingly, there were only 16 posts in Category 3 (mobilize only) and no posts placed in categories 8 (mobilize and request for feedback) and 9 (mobilize through link or quote).

Wallsten 27 FIGURE 2. Blog Use on A-list and Less Popular Political Blogs. 90% 80% 70% 60% Percent of Posts 50% 40% 30% 20% 10% 0% (1) Link or Quote Only (2) Commentary Only (3) Mobilize Only (4) Feedback Only (5) Commentary and Mobilize (6) Commentary and Feedback (7) Commentary, Mobilize and Feedback A-list Less Popular (8) Mobilize and Feedback (9) Mobilize through Link or Quote Although the vast majority of posts on less popular blogs always falls into the commentary only category, the percentage of posts that elicit feedback from readers and attempt to mobilize political action changed significantly at different moments during the campaign. Indeed, as Figure 3 shows, the percentage of posts eliciting feedback from readers spiked on the days of the three presidential debates (September 30 th, October 8 th, and October 13 th ) and on the two days immediately following the election (November 3 rd and 4 th ), while the percentage of posts attempting to mobilize political action peaked on the day of the election (November 2 nd ). Interestingly, the increased percentage of posts that mobilize political action and request feedback came at the expense of posts that link to sources without commenting rather than at the expense of posts that express opinions. More specifically, the percentage of posts classified as link or quote only dropped dramatically on the days of the presidential debates and on the days immediately following the election. It appears, therefore, that some political events such as debates and elections lead less popular political bloggers to stop using their blogs as link filters and to, instead, engage their audiences with direct calls for feedback and political action. In order to determine what kinds of political behavior bloggers attempt to encourage in their blog posts, each post that included an attempt at mobilization was also coded for the specific type of political action suggested by the blogger.

28 JOURNAL OF INFORMATION TECHNOLOGY & POLITICS FIGURE 3. Less Popular Blog Usage During the Campaign. 100% 90% 80% 70% Percent of Posts 60% 50% 40% 30% 20% 10% 0% 7/1/2004 7/15/2004 7/29/2004 8/12/2004 8/26/2004 9/9/2004 9/23/2004 10/7/2004 10/21/2004 11/4/2004 11/18/2004 Posts with Mobilization Link/Quote Only Posts with Feedback Commentary Only As Figure 4 shows, calls to sign an online petition, to participate in an online poll, to join in a boycott, to volunteer, and to attend a rally or protest event were rare on less popular political blogs. To be more precise, each of these activities was suggested in less than 5% of the mobilization posts found on less popular blogs. The two most common types of political action suggested by less popular political bloggers were contributing money to a political candidate, political party, issue advocacy organization or charity (29% of mobilization posts), and turning out to vote in the election (27% of mobilization posts). Attempts to convince readers to send e-mails to a friend, media institution, or politician were also common on less popular political bloggers, with over 23% of mobilization posts suggesting readers write an e-mail. Interestingly, media institutions were more likely to be the suggested recipients of e-mail mobilizations than either friends or politicians. Indeed, 13 of the 25 posts that called for sending e-mails suggested that readers e-mail a media institution while only 9 posts suggested that readers e-mail a politician, and only 3 posts suggested that readers e-mail their friends. While commentary only posts constituted the vast majority of less popular blog posts in the sample overall, there were some important differences in how liberals and conservatives used their blogs. To be more specific, less popular liberal

Wallsten 29 FIGURE 4. Mobilization Posts on A-list and Less Popular Blogs*. 35% 30% 25% Percent of Mobilization Posts 20% 15% 10% 5% 0% Email On-line Poll Boycott Volunteer Protest/Rally On-line Petition Contribute Money Voting Phone Call A-list Less Popular *Total adds up to more than 100% because posts can mobilize more than one kind of political action. bloggers were more than three times as likely as less popular conservative bloggers to mobilize political action (2.9% to.9%) and more than four times as likely to request feedback from their readers (3.8% to.8%). Interestingly, less popular liberal and conservative bloggers devoted a very similar percentage of posts exclusively to commentary (84% to 82%), but conservative bloggers were far more likely than liberal bloggers to pass along links or quotes without comment (19% to 10%). Political ideology, therefore, appears to have important implications for how a political blogger decides to use his or her blog. Despite the general tendency of liberal bloggers to encourage more political action and audience feedback, it is important to note that there were also large individual differences in the ways that less popular political bloggers used their blogs. Some liberal bloggers, for example, used their blogs exclusively as soapboxes and never requested feedback from their readers, attempted to mobilize political action, or passed along links or quotes without commenting on them. To be more precise, only 27 of the 67 less popular liberal blogs included in the sample attempted to mobilize political action in any of

30 JOURNAL OF INFORMATION TECHNOLOGY & POLITICS their posts, only 18 liberal blogs requested feedback from their readers in any of their posts, and only 43 liberal blogs passed along links or quotes without comment in any of their posts. Similarly, there were a number of less popular conservative bloggers who frequently encouraged political action, sought out audience feedback, and passed along information without commenting. More specifically, 10 of the 44 less popular conservative blogs included in the sample attempted to mobilize political action in at least one post, 8 conservative blogs requested feedback from their readers in at least one post, and 25 conservative blogs passed along information without commenting in at least one post. While less popular liberal bloggers were, on the whole, more likely to encourage political action and to request feedback from readers, not all liberal bloggers engaged in these activities and, more importantly, there were many conservatives who used their blogs as more than soapboxes. A-list Blogs As Figure 2 shows, A-list political bloggers used their blogs in ways that were very similar to less popular political bloggers. Indeed, the vast majority of posts on A-list blogs (81%) were commentary only posts, while approximately 13% of posts were categorized as link or quote only. A-list bloggers, like their less popular counterparts, also asked very little of their audience during the 2004 campaign. To be more specific, requests for feedback appeared in only 3% of posts, and calls to political action made up only 2.5% of A-list blog posts. Once again, Categories 3 (mobilize only), 8 (mobilize and request for feedback), and 9 (mobilize through link or quote) were sparsely populated with only three posts appearing in Category 3 and no posts appearing Categories 8 and 9. Similar to less popular political bloggers, the percentage of posts that A-list bloggers devoted to audience feedback varied at different points during the campaign. As Figure 5 shows, the percentage of posts that requested input from readers reached their peaks on the days of the presidential debates, and the overall percentage of posts devoted to audience feedback was much higher in the weeks immediately following the election than at any other point during the study. Mimicking the findings with posts on less popular blogs, requests for feedback appear to have come at the expense of posts that pass on information without commentary rather than at the expense of posts where the blogger expresses his or her opinion. As Figure 5 also shows, A-list political bloggers differed significantly from less popular political bloggers in the timing of their mobilization attempts: While less popular bloggers engaged their highest levels of mobilization in the month leading up to the election and rarely encouraged political participation after November 2 nd, A-list bloggers attempted to mobilize political action at fairly consistent rates across the period of study. Particularly revealing were the differences in posts between less popular and A-list blogs on the day of the election. Whereas less popular bloggers mobilized political action in a greater percentage of posts on Election Day than on any other day during the study (14.3% of posts), A-list bloggers encouraged readers to participate in only 4.2% of posts. Contrary to less popular bloggers, therefore, it appears that A-list bloggers tend to encourage political participation regardless of the stage in the campaign. Despite differences in timing, A-list bloggers often encouraged readers to take part in the same kinds of political activities that less popular bloggers suggested to their readers. As Figure 4 shows, calls to sign an online petition, to participate in an online poll, and to join in a boycott were as rare on A-list political blogs as they were on less popular political blogs. To be more precise, each of these activities was suggested in less than 5% of the mobilization posts found on A-list blogs. Furthermore, contributing money to a political candidate, political party, issue advocacy organization, or charity (27%), and e-mailing a friend, media institution, or politician (32%) were suggested on A-list blogs at rates that were similar to those on less popular blogs. A-list bloggers, like less popular bloggers, were more likely to encourage their readers to e-mail media institutions than to e-mail friends or politicians. More specifically, A-list bloggers suggested sending an e-mail to media institutions in 23 of the 42 posts that

Wallsten 31 FIGURE 5. A-list Blog Usage During the Campaign. 100% 90% 80% 70% Percent of Posts 60% 50% 40% 30% 20% 10% 0% 7/1/2004 7/15/2004 7/29/2004 8/12/2004 8/26/2004 9/9/2004 9/23/2004 10/7/2004 10/21/2004 11/4/2004 11/18/2004 Posts with Mobilization Link/Quote Only Posts with Feedback Commentary Only mentioned sending an e-mail, while friends and politicians were only suggested recipients in 19 posts. Although A-list and less popular bloggers spent similar amounts of time encouraging readers to sign an on-line petition, participate in an online poll, join a boycott, send an e-mail, and contribute money, there were important differences in how often they attempted to mobilize other forms of political action. In fact, A-list bloggers were much less likely to suggest voting and much more likely to suggest making a phone call than less popular bloggers. As Figure 2 shows, A- list bloggers encouraged voting in only 5% of mobilization posts (compared to 27% of mobilization posts on less popular blogs) and encouraged placing a phone call in over 18% of mobilization posts (compared to 7% of mobilization posts on less popular blogs). In short, despite mobilizing in roughly the same percentage of posts, A-list and less popular bloggers are not equally likely to suggest all forms of political action. Although every A-list blog was dominated by commentary posts, A-list liberal and conservative bloggers differed significantly in how they used their blogs: A-list liberal bloggers, like liberal less popular bloggers, were more than twice as likely as A-list conservative bloggers to mobilize political action (3.5% to 1.7%) and more than four times as likely to request feedback from their readers (4.9% to 1.2%). Interestingly, A-list liberal and A-list conservative bloggers provided commentary in approximately the same percentage of posts (81.9% to 80.9%), but conservative bloggers were far more likely than liberal bloggers to pass along

32 JOURNAL OF INFORMATION TECHNOLOGY & POLITICS links or quotes without comment (16.1% to 9.8%). Similar to less popular bloggers, therefore, ideology appears to influence what kinds of things an A-list political blogger does with his or her blog. Despite the fact that liberal bloggers were, overall, more likely to mobilize political action and to request feedback from their readers, there were significant individual differences in how A-list political bloggers used their blogs. Liberal blogs Talking Points Memo (http://www.talkingpointsmemo.com), Informed Comment, and Tapped (http://www.prospect.org/weblog), for example, attempted to mobilize political action in less than 1% of posts, while conservative blogs Winds of Change (http://www.windsofchange.net), Michelle Malkin (http://www.michellemalkin. com), and The Indepundit (http://www.indepundit. com) suggested that their readers engage in some form of political participation in over 4% of posts on their blogs. Similarly, while liberal blogs Hullabaloo (http://www.digbysblog.blogspot. com), Informed Comment, and Matthew Yglesias (http://www.yglesias.typepad.com) rarely asked their readers for input, conservative blogs Winds of Change and Little Green Footballs (http:// www.littlegreenfootballs.com) made reader feedback a key component of their blogs. Although liberal A-list political bloggers as a group showed a greater willingness to use their blogs for more than just opinionated commentary, there were significant differences in how each individual blogger decided to use his or her blog. DISCUSSION What do these results suggest about the ways that political bloggers actually use their blogs? First, political bloggers use their blogs primarily as soapboxes and, to a lesser extent, as transmission belts. More precisely, the vast majority of posts overall (95.7%), and the vast majority of posts on every blog studied here, were used to make opinion statements or to pass on information without comment and not to mobilize political action. While these findings should not be totally surprising, given recent survey data that suggests political bloggers blog because they want to express their political opinions and provide readers with new sources of information (McKenna & Pole, 2004), they do imply that political bloggers place less emphasis on directly influencing the institutionalized political process than they do on changing the prevailing political discourse. In other words, it appears that the primary goal of political bloggers is to persuade and inform their audiences. Second, in the rare instances when political bloggers attempt to mobilize political action, they are likely to encourage readers to contact politicians and media organizations. Specifically, 22% of mobilization posts on A-list blogs and less popular blogs suggested e-mailing or calling a media organization, and 14% of mobilization posts suggested e-mailing or calling a politician. Although these findings do not paint a complete picture of the relationship that political bloggers have with journalists and political elites, they are consistent with the frequently heard claim that bloggers adopt a critical if not confrontational stance toward mainstream media coverage and statements by politicians. Indeed, every one of the posts that encouraged readers to contact a media organization or politician insisted that readers express their dissatisfaction with the way a particular issue was being discussed. To state all of this more simply, political bloggers may very well be using their blogs to act as a fifth estate (Cornfield, Carson, Kalis, & Simon, 2005) or watchdog that checks the mainstream media and political elites (McKenna & Pole, 2004). Finally, while political bloggers are generally reluctant to request feedback from their readers, they do seek out the opinions of their audiences in response to major political events. In particular, requests for feedback increased dramatically on the days of the presidential debates and in the weeks immediately following the election. Although the findings presented here do not assess the extent to which requests for feedback are actually heeded by blog readers, they do suggest that the level of political dialogue between bloggers and their audiences is not constant over time. In other words, blogs do provide citizens with the

Wallsten 33 opportunity to engage in dialogue with each other, but it appears that political bloggers are more likely to capitalize on that opportunity during some times rather than others. CONCLUSIONS As the number of blogs has grown exponentially over the last five years, so have the number of studies that seek to assess their political significance. This paper attempted to contribute to this burgeoning literature by asking the question: how do political bloggers actually use their political blogs? Consistent with previous research that finds political bloggers use their blogs to express their political beliefs, to interact with like-minded people, to inform their readers, and to influence the political world around them (McKenna, 2007; McKenna & Pole, 2004; McKenna & Pole, in press), this paper has shown that political blogs are complex forms of political participation that contain a mix of opinion statements, mobilization attempts, requests for audience feedback, and links to information produced by others. The findings presented here also suggest, however, that these ingredients are not mixed evenly and that their relative importance varies across time and across bloggers. More specifically, although political bloggers use their blogs primarily as soapboxes, blog use changed significantly in response to key political events (such as presidential debates and Election Day). In addition, liberal bloggers regardless of popularity were far more likely to mobilize political action and request feedback from their readers than conservatives. Political blogging, in short, is a complex form of political participation that blends hypertext links, opinionated commentary, calls to political action, and requests for feedback in different ways at different moments in time. The findings presented here suggest a number of directions for future research. First, a more detailed content analysis than the one presented here is needed in order to better assess the nature and substance of the political commentary found on political blogs. In particular, future research should explore the agendas and issue frames found on political blogs in order to test the increasingly popular assertion that blogs act as a fifth estate or watchdog for the mainstream media and political elites. While this paper provides initial evidence that bloggers attempt to check mainstream media and political leaders by showing that a large percentage of the mobilization attempts made by political bloggers are directed against these actors, the question of whether blogs act as watchdogs or as echo chambers cannot be adequately answered until more is known about the extent to which bloggers fact check, criticize, and contextualize the statements of journalists and politicians. Second, future research should explore the extent to which political bloggers are actually engaged in dialogue with their readers. Although the findings presented here suggest that dialogue between bloggers and readers is rare, an alternative measure of dialogue that does not rely solely on explicit requests for audience feedback may paint a much different picture: Many blogs contain comments sections that researchers may analyze in order to assess whether bloggers are receiving feedback from their audiences even in the absence of explicit requests for reader input. If reader participation in comments sections is found to be robust, the conclusions presented here may dramatically underestimate the amount of dialogue taking place in the political blogosphere. Third, future research should test the findings presented here by exploring political blog use during different periods of time. Although this paper has suggested that political blogs function primarily as soapboxes, it is possible that these findings are contingent upon the unique circumstances of the 2004 campaign. Given that most mobilization attempts on the political blogs in the sample centered on voting and campaign contributions, for example, there is a good chance that political blogs will look even more like soapboxes when an election is not looming on the horizon. Similarly, increasing contact with the major political parties (Bluey, 2006; Pfeiffer, 2006), growing levels of readership, and a changing political context may lead political bloggers to provide less commentary and more mobilization.