Twitter Use by Presidential Primary Candidates During the 2012 Campaign

Abstract

This study examines Twitter use by presidential candidates during the 2012 primary election. The Twitter feeds and activity levels of candidates from the Republican, Democratic, Libertarian, and Americans Elect parties and their campaigns were gathered over a 3-month span. Variables examined include the number of tweets posted, followers gained, and followers added as well as the occurrence of hashtags, user mentions, hyperlinks, and content categories within tweets. Results showed candidates’ presence on Twitter was not uniform. Tweet frequency did not necessarily result in followers gained. Interestingly, the candidates who tweeted the most were not major party nominees. As the Republican primary campaign progressed, the amount of daily tweeting by the candidates was not higher than it had been earlier in the primary season.

Keywords

Twitter presidential candidate presidential primary

Technological advances in communication accompany every campaign cycle. In an attempt to reach citizens and media outlets effectively, candidates make use of technologies in innovative ways. As politicians try to replicate the touted “Obama model” (Towner & Dulio, 2012), placing their stock in the land of Web 2.0, it is important to evaluate the role of online media in the so-called “presidential package.” Analyzing the use of social networking sites (SNSs), especially Twitter, is therefore key to understanding the political terrain of the 2012 presidential campaign.

The current study addresses the following questions: To what extent did the presidential candidates adopt and implement advanced Twitter strategies in 2012, and did the major party candidates have a different style and presence on Twitter than did other candidates? To answer these questions, we examined the Twitter feeds of the Republican, Democratic, Libertarian, and Americans Elect presidential primary candidates and their campaign feeds between February 1 and May 2, 2012. An analysis of post, hyperlink, user mentions, and hashtag frequencies, followers gained and added, and a preliminary look at popular Twitter content categories revealed that the candidates’ presence on Twitter was not uniform. Moreover, the candidates who tweeted the most were not major party candidates.

Challenges and Benefits of Twitter Use

After the 2008 election, scholars predicted increased and targeted web use by political campaigns at the federal and local level (Towner & Dulio, 2012). This included use of SNSs, which allow candidates to build profiles and showcase connections within a delimited system (boyd & Ellison, 2007). Among these sites, Twitter is unique due to its confinement to 140-character messages and the lack of restrictions on viewing messages in the form of account-owner permissions. Followers are said to establish connections for the content, rather than the relationships, resulting in numerous ties that span physical and social disparities (Virk, 2011). Twitter brings with it new possibilities for candidate-voter interaction as the “@username” function allows candidates to reply directly to other users and promote dialogue. Managing a Twitter audience, therefore, requires constant activity to respond to, monitor, and understand audience interests (Marwick & boyd, 2011).

It is not surprising that this platform is fraught with potential communication challenges. These challenges include an absence of authoritative hierarchies (Metzgar & Maruggi, 2009), the possible loss of message control (Gueorguieva, 2008; Johnson & Perlmutter, 2010; Stromer-Galley, 2006), and overall blurring of traditional audience conceptualizations (Marwick & boyd, 2011). Scholars and pundits also question whether the overall use of SNSs by politicians actually matters when it comes to voting outcomes (Baumgartner & Morris, 2010; Ferenstein, 2012; Kushin & Yamamoto, 2010; Zhang, Johnson, Seltzer, & Bichard, 2010). Although the number of Twitter users continues to increase, only a fraction of those users report using the site to gather political information (Smith, 2011; Smith & Rainie, 2008). Right now, Twitter and other SNSs are still seen as complementary to traditional outreach mediums (Towner & Dulio, 2012). The true payback may be in organizing volunteers and activists, an aspect some maintain is overlooked (Abroms & Lefebvre, 2009; Sifry, 2012).

The primary benefits of the SNS as a campaign tool are said to include low cost, enhanced recruitment of volunteers and contributions, and an opportunistic space for lesser known candidates (Gueorguieva, 2008). One benefit of all social media is the direct and unrestrained nature of the communication and the bypassing of traditional media outlets (Lassen & Brown, 2011). Such direct forms of marketing present opportunities to enhance support systems and facilitate meaningful connections with constituents (Ansolabehere, Snyder, & Stewart, 2001). In addition to the aforementioned benefits, Twitter is touted for its brevity and frequency (Java, Song, Finin, & Tseng, 2007). Public relations professionals suggest that Twitter is well-suited for location-based advertising, guerilla marketing, and the solicitation of micro-payments (Virk, 2011). Sophisticated Twitter strategies, however, are rarely seen below the federal level, with the Obama campaign continuing to set the bar (Abroms & Lefebvre, 2009; Towner & Dulio, 2012).

Candidate Blog, Microblog Strategies

Prior to the launch of Twitter in 2006 (labeled a microblogging site due to its character limit), candidates did not display a cohesive strategy in the political blogosphere. Though hyperlinks were extremely common across blog posts, candidates and their staff members lacked an understanding of its primary function, made evident by disparate approaches (Lawson-Borders & Kirk, 2005) and a lack of cohesion between blogs and websites (Williams, Trammell, Posteinicu, Landreville, & Martin, 2006). Due to the relative infancy of SNSs, this trend may have followed over into the social media arena.

Existing literature on Twitter use by politicians is limited in that most academic research of Twitter feeds focuses on use by members of the House and Senate (Golbeck, Grimes, & Rogers, 2010; Lassen & Brown, 2011). In the current study, we expand this line of research to presidential primary candidates. An examination of Twitter use by members of Congress found that they rarely promote interaction through “re-tweets” or user mentions. Instead, they prefer to use the site to disseminate the same information shared in other media (Golbeck et al., 2010). Such displays imply that candidates are still trying to use Twitter in a top-down manner. Most congresspersons also fail to post regularly or to keep readers up to date. Tweets focus overwhelmingly on information (e.g., facts, opinions, issues, and news), followed by posts concerning activities such as speeches and appearances. The presence of links was common, appearing in approximately 45% of tweets, with the majority referencing news websites and blog posts by individual congresspersons, followed by links to personal websites and web pages of political action committees and partisan organizations. In a study of Twitter adoption and use by members of both chambers, Lassen and Brown (2011) found it difficult to identify the actual motivation behind Twitter adoption and regular use, with age, Republican membership in the House, and Republican leadership being the only associated variables. Republican minority status was identified as a possible impetus behind greater Twitter use by party members; however, very few electoral factors played a role. The association between Twitter use and leadership echoes findings within the marketing sector that suggest those occupying influential roles are more likely to see the benefits behind adoption of social media (Sweetser & Kelleher, 2011).

In terms of SNS use in presidential campaigns, so far, researchers have examined presidential candidate Facebook profiles (Compton, 2008) and YouTube use (Church, 2010), as well as overall Twitter network activity during the 2008 presidential election (Metzgar & Maruggi, 2009). Analysis of Twitter use by the president and presidential candidates is limited, however. In a study of Twitter use by President Obama, UK Prime Minister David Cameron, and Israeli Prime Minister Benjamin Netanyahu, Aharony (2012) found that President Obama posted the most tweets and had the most followers of the three politicians. Information about daily happenings and top issues composed the largest category of tweets for all leaders. For Obama, this was followed by quotations concerning decisive topics. During the 2008 election, Abroms and Lefebvre (2009) found that Obama’s Twitter followers outnumbered McCain’s 24 to 1 by the end of the 2008 election cycle. Tracking candidates’ online presence during the 2009 Romanian presidential election, Aparaschivei (2011) measured the use of Twitter over a 2-and-a-half-month period. By tracking the number of tweets, followers, and following, the study revealed that candidate use of Twitter varied greatly, once again exhibiting a lack of strategy.

The current study seeks to discover whether a consistent implementation of Twitter use by presidential primary candidates in 2012 occurred. Inconsistent use of Twitter across the 2012 primary candidates suggests politicians are still having difficulties using this new online medium. On the other hand, if Twitter use is relatively uniform, it suggests candidates, at least at the federal level, are becoming more familiar with the components and best practices in use for this online media platform. To answer our research questions, we analyzed Twitter use by presidential candidates during the 2012 primary election, including activity levels; use of hyperlinks, hashtags, and the “@username” function; and emphasis on certain content by American politicians.

Method

Data on Twitter activity by the 2012 Republican, Democratic, Libertarian, and Americans Elect presidential primary candidates were collected over a 3-month span. Computer-assisted content analysis was used to determine popular content categories within posts, and descriptive statistics were used to describe overall candidate activity levels.

Data Collection

Tweets posted to individual candidate Twitter feeds and their campaign feeds were gathered between February 1 and May 2, 2012. During this time period, Republican candidate Rick Santorum resigned from his campaign on April 10, 2012. On May 2, 2012, Republican Newt Gingrich announced his withdrawal, solidifying Romney’s bid and marking the end of our data collection for this study. All Twitter feeds analyzed in this project were verified by Twitter. They included those of Republican candidates Fred Karger, Mitt Romney, Newt Gingrich, Rick Santorum, and Ron Paul; Democratic candidate Barack Obama; Americans Elect candidate Buddy Roemer; and Libertarian candidate Gary Johnson. Though others were involved in the election early on (e.g., Michele Bachmann, Jon Huntsman, etc.), only candidates in the race at the start of data collection were included. Campaign feeds included ElectRoemer, Gingrich2012HQ, and TeamRomney. Tweets were gathered by downloading Twitter html data on a weekly basis. These data were then entered into an Excel spreadsheet, displaying the user’s account name, the date a tweet occurred, and its complete content. These feeds did not contain re-tweets unless the user copied the message and posted it himself or herself. Tweets were later examined for use of web addresses, hashtags, and @username references. To understand activity levels among the different candidates, the number of tweets posted, the number of followers an account holder accrued, and the number of people an account holder chose to follow were gathered on a nightly basis. Data collection occurred nightly between 11 p.m. and midnight.

Coding Procedures

For the purpose of establishing popular content categories, the unit of analysis was the individual reference to any of the four content categories examined across tweets posted by a given candidate or campaign during the entire study period. QDA Miner, a computer-assisted coding program, and its companion program WordStat, an advanced text mining software tool, were used to code the number of references falling under specific content categories. Categories were similar to those used by Golbeck et al., (2010). Dictionaries were created to determine references to the following categories:

Issues: Any references to important election issues such as the economy, health care, or foreign affairs.¹

Campaign-related media: Any references to campaign-related media, including advertisements, endorsements, and polls.

Events: Any references to campaign events such as rallies, speeches, or debates.

Calls for action: Any references to support given or needed including donations, volunteers, or votes.

Analysis

We did not test for statistical differences between the candidates as our data are based on population figures, not projections. Researchers sometimes mistakenly use p values to connote substantive associations or differences, which of course p values cannot do. Any differences shown between the candidates are real differences in this study. The extent to which the differences are meaningful substantively is a matter of interpretation.

Results

The candidates had different levels of Twitter involvement coming into the primaries. Figure 1 shows the number of lifetime tweets the candidates had made to date on February 1, 2012, and May 2, 2012. Although Obama’s Twitter feed was created well in advance of the other primary candidates (Obama profile created March 5, 2007), he was not the most frequent tweeter among them. Buddy Roemer (profile created February 28, 2011) had tweeted 5,823 times by February 1, and his campaign feed ElectRoemer (profile created May 16, 2011) had put out 3,722 tweets by that time. The next most frequent tweeters were New Gingrich (profile created February 12, 2009) at 3,190 tweets and his campaign feed Newt2012HQ (profile created February 24, 2011) at 5,289 tweets. Other Twitter feeds were created as follows: Karger on August 20, 2008, Johnson on December 9, 2009, Romney on June 23, 2009, Santorum on July 20, 2009, Paul on April 24, 2011, and TeamRomney on November 12, 2010. As shown in Figure 1, Romney, the eventual Republican nominee, did not tweet as much as many of the other candidates; his combined candidate and campaign feeds totaled 1,455 tweets, which was under Santorum’s 1,557 tweets on February 1. Romney’s combined candidate and campaign totals, however, did outpace Paul’s 803 tweets.

Figure 1.

Number of lifetime tweets made by the 2012 candidates and campaigns to date on February 1 and May 2, 2012.

Figure 2 and Figure 3 display the number of followers the feeds had and the number of profiles that the accounts were following. Obama’s numbers are not presented in these graphs because his numbers were so large that they obscured the differences that can be seen between the other candidates. On February 1, Obama had 12,272,966 followers, which was over 8.5 times more than Gingrich’s 1,422,135 followers and was over 39 times more than Romney’s 313,140 followers. In terms of Twitter profiles that the accounts were following on February 1, Obama was following 682,926 profiles, in comparison to Buddy Roemer, who was following 2,640 profiles; ElectRoemer, which was following 3,720 profiles; and Newt2012HQ, which was following 4,164 profiles. Romney, however, was following only 235 profiles, and his campaign feed TeamRomney was following 248 profiles.

Figure 2.

Number of Twitter followers of the 2012 candidates and campaigns on February 1 and May 2, 2012.

Figure 3.

Number of Twitter profiles being followed by the candidates and the campaigns on February 1 and May 2, 2012.

We examined the correlations between tweet frequencies, the numbers of followers, and the numbers of profiles being followed by the candidates and the candidate campaign accounts (n = 11). The results indicated that there was little relationship between candidates’ tweet frequencies and the number of followers they had (r = .024 for February 1; r = .029 for May 2) or between tweet frequencies and the number of profiles being followed (r = .031 for February 1; r = .055 for May 2). There was, however, a strong linear association between the number of followers and the number of profiles being followed by the candidates and campaigns (r = .993 for February 1; r = .995 for May 2).

Candidate activities on Twitter between February 1 and May 2, when Gingrich resigned from the campaign, revealed disparate levels of involvement on behalf of the Republican presidential hopefuls and their campaigns, as shown in Table 1. Buddy Roemer with 2,052 and ElectRomer with 2,502 total tweets made between February 1 and May 2 outpaced the other candidates, save for Obama. Even though Santorum dropped out of the campaign on April 10, he tweeted during the period under investigation more than did Romney but less than did Gingrich. Despite tweeting less than Roemer and ElectRoemer, Romney acquired 169,664 followers, Gingrich obtained 32,921 followers, and Santorum gained 102,272 followers during this time. While TeamRomney was less active in terms of tweeting than was Gingrich2012HQ, TeamRomney gained more followers.

Table 1.

Candidate and Campaign Twitter Use Totals and Averages from February 1 to May 2, 2012.

	Barack Obama	Buddy Roemer	Fred Karger	Gary Johnson	Mitt Romney	Newt Gingrich	Rick Santorum	Ron Paul	Elect Roemer	Gingrich 2012HQ	Team Romney
Total tweets/re-tweets	995	2,052	459	193	132	385	339	144	2,502	1,354	203
Total followers gained	2,627,568	7,784	323	6,432	169,664	32,921	102,272	65,861	5,919	821	2,251
Total profiles followed	–4,296	411	14	217	12	–18	33	3	6,588	1,466	99
Mean (SD) tweets/re-tweets per day	10.8 (5.5)	22.3 (15.4)	5.0 (5.5)	2.1 (1.8)	1.4 (1.1)	4.2 (3.7)	3.7 (2.9)	1.6 (1.5)	27.2 (17.5)	14.7 (14.6)	2.2 (4.6)
Mean (SD) followers gained per day	28,560.5 (14,782.7)	84.6 (44.8)	3.5 (4.8)	69.9 (149.5)	1,844.2 (979.9)	357.8 (1,644.8)	1,111.7 (1,847.2)	715.9 (347.6)	64.3 (54.1)	8.9 (36.5)	24.5 (39.7)
Mean (SD) new profiles followed per day	–46.7 (153.7)	4.5 (5.0)	0.2 (1.2)	2.4 (16.2)	0.1 (0.5)	–0.2 (2.5)	0.4 (0.9)	0.0 (0.2)	71.6 (441.5)	15.9 (139.7)	1.1 (3.1)

One way a candidate can presumably cultivate a following is by following others on Twitter. As shown in Table 1, Roemer added 411 profiles to the list of feeds he was following and ElectRoemer added 6,588 between February 1 and May 2. Gingrich’s net number of profiles he was following decreased by 18 during the time studied, but his campaign feed added 1,466 profiles to the list it was following. Neither Romney nor TeamRomney made much use of profile following, with only 12 and 99 profiles added, respectively. Obama downsized the number of profiles that he was following by 4,296 during the time period under investigation. This did not appear to hurt his ability to gain followers as his following increased by 2,627,568 people, which meant that he acquired 6.6 times the following of the other 10 candidates and campaign feeds tracked in this study combined.

We investigated the dynamics of the Republican tweet and follower frequencies between February 7 and May 2 to see if there appeared to be an increase in tweets and followers gained per day as the nomination campaign wore on. As shown in Figure 4, which presents the daily tweet frequencies of the Republican candidates combined with their campaign feed totals using 7-day prior moving averages, there are ebbs and flows to the tweet frequencies, but it does not appear that candidates/campaigns made more use of Twitter as the campaign progressed. It appears that Gingrich/ Gingrich2012HQ made the most use of Twitter prior to Super Tuesday, but then the number of daily tweets decreased. Figure 5 shows that as the campaign continued, Romney’s following increased on a daily basis. Gringrich, Paul, and Santorum had jumps in their daily gains in following in late March, but then their daily gains turned into losses. Once Santorum, for example, dropped out of the campaign, people stopped following his feed.

Figure 4.

Number of tweets per day by the Republican candidates/campaigns between February 7 and May 2, 2012 (7-day prior moving average).

Figure 5.

Number of followers gained per day by the Republican candidates/campaigns between February 7 and May 2, 2012 (7-day prior moving average).

In terms of the content of tweets posted by individual candidates—excluding those re-tweets that are not present in html feeds—all candidates posted links to other websites, mentioned other users, and included hashtags. Across candidates, 54.5% of tweets contained links to other media, 43.7% of tweets mentioned other users (@username), and 40.4% of tweets referenced one or more hashtags. Use of these options by candidate is shown in Table 2. Interestingly, Paul made the greatest use of both web addresses and hashtags and used the @username option the least. A look at his tweets revealed that they were quite uniform, featuring a comment, followed by a related website, followed by a related hashtag. Johnson, Gingrich, Santorum, and Paul all used web addresses more than the other two options, while Karger and Roemer made greater use of the @username function, showing a propensity to create dialogue with others. It is important to note, however, that Johnson used the “@username” function more than any other candidate, Romney employed this option regularly, and Karger and Roemer used all three options to a similar extent. Only Gingrich, Santorum, and Paul chose to use hashtags more than references to other users. Obama’s use of the three options somewhat mirrored that of Gingrich, with each using more web addresses than hastags, followed by user mentions. In terms of categories referenced across all tweets, the majority focused on important issues facing the country. Of the references coded by Wordstat, 50.5% were made about election issues, 25.5% referenced campaign support either already given or needed in the future, 12.7% referenced campaign events and candidate appearances, and 11.4% referenced candidate media.

Table 2.

Percentage of Tweets Containing Web Addresses, Profile Mentions (@username), and Hashtags.

	Barack Obama	Buddy Roemer	Fred Karger	Gary Johnson	Mitt Romney	Newt Gingrich	Rick Santorum	Ron Paul
Website links	77.6	40.5	62.6	86.1	51.1	68.8	62.3	96.4
User mentions	26.0	47.4	63.3	65.5	48.1	24.2	25.6	7.2
Hashtags	44.9	39.8	32.7	24.2	22.6	55.1	28.1	97.8

Note: Percentages for website, hashtag, and @username usage are based on whether each occurred within an individual tweet. Percentages do not reflect use of such items within re-tweets, only original candidate posts.

Discussion

This study showed that Twitter was not employed consistently by the 2012 presidential primary candidates. There were candidate differences in the frequency of tweeting. Candidates came into the primary season having established their Twitter accounts in different years and having different lifetime tweet totals. Interestingly, the number of tweets made by a politician was not strongly associated with how many followers the politician and/or his campaign had acquired. Between February 1 and May 2, 2012, there was not a consistent increase in tweeting frequency across the candidates as the campaign progressed. Gingrich, Romney, and Santorum increased their amounts of tweeting as they approached Super Tuesday. Their levels of tapering off in frequency differed.

The results of this study are similar to previous studies in that most politicians are not using Twitter to post regular, timely updates (Aparaschivei, 2011; Golbeck et al., 2010; Lassen & Brown, 2011). If being on Twitter is “like being delivered a newspaper whose headlines you’ll always find interesting,” candidates are not yet treating it as such (www.twitter.com). With news outlets regularly using blogs as legitimate news sources (Messner & Distaso, 2008), candidates could take advantage by microblogging in a way that attracts the attention of news outlets. The heavy use of hyperlinks suggests, at least currently, they may be following rather than leading.

On the other hand, “there is no ‘right’ or ‘wrong’ way to use Twitter” (Golbeck et al., 2010). As suggested by the current results, candidates did not necessarily see payoff equal to their Twitter use. Though Buddy Roemer tweeted more than any other candidate, his increase in followers did not come close to that of other well-known candidates. This finding suggests that the popularity on Twitter is influenced more by forces outside the network than the frequency of posting. As an Americans Elect candidate, Roemer was not as recognizable as those from the Republican and Libertarian parties. Early on, when running as a Republican, Roemer was outspoken about being excluded from televised debates (Henderson, 2011). The fact that his campaign feed was able to use Twitter to attract the attention of supporters gives weight to the stipulation that online platforms may be helpful in bringing attention to lesser known candidates (Gueorguieva, 2008).

Though candidates regularly mentioned other Twitter users in their posts, the question remains as to whether this resulted in actual “meaningful dialogue.” Not all user mentions were directed toward voters, with many being used to call out other candidates or the president. For example, in many instances Romney specifically targeted @BarackObama in his tweets. This was not done to create dialogue with constituents but was done to criticize the opposition. Future research should examine the extent to which candidates are interacting with one another on Twitter, as well as the tone and content of such conversations. Though social media may be used to reach voters, especially young voters, it will do little if the information provided is uninformative (Baumgartner & Morris, 2010; Lassen & Brown, 2011) or overwhelmingly negative.

As we look to the next presidential campaign cycle, we anticipate that candidates will continue to use Twitter as long as it remains a popular SNS platform. By that point, there may be established best practices of its effective use. It may be the case that future candidates will figure out innovative ways to mobilize supporters on Twitter as President Barack Obama has done during his candidacies. Alternatively, it may be the case that candidates merely engage in Twitter activity pro forma as an expectation put forth by the news media, which tends to question whether candidates are in touch with voters if candidates do not have a presence on the most popular SNS platforms of the day, regardless of whether that presence actually matters to the voters. If the latter is the case, then future studies will most likely reflect the same results as this one, which indicates candidate presence but inconsistency in use and content.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

Author Biographies

Bethany Anne Conway is a doctoral student in the Department of Communication at the University of Arizona. Her research interests involve mass media impact on public opinion and government policy with a focus on the changing news media environment.

Kate Kenski, PhD, is an associate professor in the Department of Communication and School of Government & Public Policy at the University of Arizona, where she teaches political communication, public opinion, and research methods. She has published over 30 articles, books, and book chapters on political communication. Her book The Obama Victory: How Media, Money, and Message Shaped the 2008 Election (coauthored with Bruce W. Hardy and Kathleen Hall Jamieson; June 2010, Oxford University Press) has won several awards including the PROSE Award for 2010 Best Book in Government & Politics, the 2011 International Communication Association Outstanding Book Award, the 2012 National Communication Association Diamond Anniversary Book Award, and the 2012 National Communication Association Political Communication Division Roderick P. Hart Outstanding Book Award.

Di Wang is a doctoral candidate in the Department of Communication at the University of Arizona. She is going to be an assistant professor in the Faculty of Humanities and Arts at Macau University of Science and Technology in September, 2013. Her scholarly interests include political communication, selective exposure and perception, and new media.

References

Abroms

L. C.

Lefebvre

R. C.

(2009). Obama’s wired campaign: Lessons for public health communication. Journal of Health Communication, 14, 415–423. doi: 10.1080/10810730903033000

Aharony

(2012). Twitter use by three political leaders: An exploratory analysis. Online Information Review, 36, 587–603. doi:10.1108/14684521211254086

Ansolabehere

Snyder

J. M.

Stewart

(2001). Candidate positioning in the U.S. House elections. American Journal of Political Science, 45, 139–159.

Aparaschivei

P. A.

(2011). The use of new media in electoral campaigns: Analysis of the use of blogs, Facebook, Twitter, and YouTube in the 2009 Romanian presidential campaign. Journal of Media Research, 2, 39–60.

Baumgartner

Morris

J. S.

(2010). MyFaceTube politics: Social networking web sites and political engagement of young adults. Social Science Computer Review, 28, 24–44. doi:10.1177/0894439309334325

boyd

d. m.

Ellison

N. B.

(2007). Social network sites: Definition, history, and scholarship. Journal of Computer-Mediated Communication, 13, 210–230. doi:10.1111/j.1083-6101.2007.00393.x

Church

S. H.

(2010). YouTube politics: YouChoose and leadership rhetoric during the 2008 election. Journal of Information Technology & Politics, 7, 124–142. doi:10.1080/19331681003748933

Compton

(2008). Mixing friends with politics: A functional analysis of ’08 presidential candidates social networking profiles. Paper presented at the annual meeting of the National Communication Association, San Diego, CA.

Ferenstein

(2012, February 26). Waiting for the “Twitter election”? Keep waiting. Advertising Age. Retrieved February 3, 2013, from http://adage.com/article/digital/waiting-twitter-election-waiting/232965

10.

Golbeck

Grimes

J. M.

Rogers

(2010). Twitter use by the U.S. Congress. Journal of the American Society for Information Science and Technology, 61, 1612–1621. doi: 10.1002/asi.21344

11.

Gueorguieva

(2008). Voters, MySpace, and YouTube: The impact of alternative communication channels on the 2006 election cycle and beyond. Social Science Computer Review, 26, 288–300. doi:10.1177/0894439307305636

12.

Henderson

(2011, October 11). Buddy Roemer: Let me participate in the Republican debate. The Washington Post. Retrieved February 6, 2013, from http://www.washingtonpost.com/blogs/post-politics/post/buddy-roemer-let-me-in-the-debate/2011/10/11/gIQAoU7tcL_blog.html

13.

Java

Song

Finin

Tseng

(2007). Why we twitter: Understanding microblogging usage and communities. In Proceedings of the 9th WebKDD and 1st SNA-KDD 2007 Workshop on Web Mining and Social Network Analysis, San Jose, CA.

14.

Johnson

T. J.

Perlmutter

D. D.

(2010). Introduction: The Facebook election. Mass Communication and Society, 13, 554–559. doi:10.1080/152005436.2010.517490

15.

Kushin

M. J.

Yamamoto

(2010). Did social media really matter? College students’ use of online media and political decision making in the 2008 election. Mass Communication & Society, 13, 608–630. doi:10.1080/15205436.2010.516863

16.

Lassen

D. S.

Brown

A. R.

(2011). Twitter: The electoral connection? Computer Review, 29, 419–436. doi:10.1177/0894439310382749

17.

Lawson-Borders

Kirk

(2005). Blogs in campaign communication. American Behavioral Scientist, 49, 548–559. doi:10.1177/0002764205279425

18.

Marwick

A. E.

boyd

(2011). I tweet honestly, I tweet passionately: Twitter users, context collapse, and the imagined audience. New Media & Society, 13, 114–133. doi:10.1177/1461444810365313

19.

Messner

Distaso

M. W.

(2008). The source cycle: How traditional media and weblogs use each other as sources. Journalism Studies, 9, 447–463. doi:10.1080/14616700801999287

20.

Metzgar

Maruggi

(2009). Social media and the 2008 U.S. presidential election. Journal of New Communications Research, 4, 141–165.

21.

Sifry

M. L.

(2012, February 28). How social media is keeping the GOP primary going. TechPresident. Retrieved February 6, 2013, from http://techpresident.com/news/21841/how-social-media-keeping-gop-primary-going

22.

Smith

(2011). Why Americans use social media. Pew Internet & American Life Project. Retrieved February 6, 2013, from http://www.pewinternet.org/Reports/2011/Why-Americans-Use-Social-Media/Main-report.aspx

23.

Smith

Rainie

(2008). The Internet and the 2008 election. Pew Internet & American Life Project. Retrieved September 1, 2011, from http://pewinternet.org/~/media//Files/Reports/2008/PIP_2008_election.pdf.pdf

24.

Stromer-Galley

(2006). On-line interaction and why candidates avoid it. Journal of Communication, 50(4), 111–132. doi:10.1111/j.1460-2466.2000.tb02865.x

25.

Sweetser

K. D.

Kelleher

(2011). A survey of social media use, motivation, and leadership among public relations practitioners. Public Relations Review, 37, 425–428. doi:10.1016/j.pubrev.2011.08.010

26.

Towner

T. L.

Dulio

D. A.

(2012). New media and political marketing in the United States: 2012 and beyond. Journal of Political Marketing, 11, 95–119. doi:10.1080/15377857.2012.642748

27.

Virk

(2011). Twitter: The strength of weak ties. University of Auckland Business Review, 13(1), 19–21.

28.

Williams

A. P.

Trammell

K. D.

Posteinicu

Landreville

K. D.

Martin

J. D.

(2006). Blogging and hyperlinking: Use of the Web to enhance viability during the 2004 US Campaign. Journalism Studies, 6, 177–186.

29.

Zhang

Johnson

T. J.

Seltzer

Bichard

S. L.

(2010). The revolution will be networked: The influence of social networking sites on political attitude and behavior. Social Science Computer Review, 28, 75–92. doi:10.1177/0894439309335162