Framing News in 140 Characters: How Social Media Editors Frame the News and Interact with Audiences via Twitter

Abstract

The social media editor, the newsroom liaison to the digital world, is the newest position in journalism. This study used content analysis to examine how these editors interacted with audiences on Twitter; which frames occurred within the news articles they posted via Twitter and if these two points of foci varied according to the media format the SMEs represent--print vs. TV news. TV SMEs were more personal in their interaction with their Twitter followers. They also posted more articles that carried the technology frame while print SMEs emphasized the human interest, conflict and economic impact frames the most. Overall, all SMEs emphasized technology and human interest stories while downplaying the conflict and economic impact frames. This particular finding goes against the norm since research shows that mainstream news coverage emphasizes conflict and economic frames while readers follow disaster, economic and political news the closest.

Introduction

Few words better capture journalism's latest development than Stephanie Gleason's in an article she penned for the American Journalism Review. "Trending in journalism right now: #social media editors" (Gleason, 2010: 6).The hash tag she uses is of course not an accidental reference to Twitter's popular method of defining keywords and trending topics. Social media editors (SMEs) can be loosely defined as tech savvy journalists appointed by news agencies to act as the newsroom liaisons to the digital world, also called the social mediasphere. They go by different monikers such as: Social media editor, social media strategist, digital editor, social media manager, social media producer, digital news editor, etc. While this position is still in its infantile stage as evidenced by the lack of consensus even on title descriptions, the SMEs' main purpose is to connect the media agency to the digital world.

Specifically, their main functions are: to monitor the social media sphere for latest trends; to procure relevant material for editors; to maintain the organization's online presence; to interact with the readers and to post news articles online. Such editors include Lauren McCullough of the Associated Press, Etan Horowitz of CNN International, and Tony Pierce of the Los Angeles Times among many others. This paper examines how the SMEs perform the last two functions: audience interaction and posting news articles online.

Why Twitter?

Three reasons justify why Twitter in particular is worth studying as far as both SMEs and new media journalism are concerned:

1) Twitter's uniqueness and suitability for audience interaction: While SMEs use different social media platforms in their reporting activities, Twitter is probably best suited for interaction purposes. Its uniqueness lies in the ability it gives an SME to reach multitudes of dedicated readers in a short time and with pithy bursts of information. Twitter allows only 140 characters per tweet and users have to write posts that deliver a quick and effective message. Additionally, SMEs (and other Twitter users) can amass large groups of dedicated readers called followers, who subscribe to a specific SME's Twitter feed because of a specific and non-temporal reason. This guarantees a long lasting audience for an SME. In turn, the SME keeps followers engaged by posting tweets at a steady pace (sometimes several times per hour), thus cementing the relationship. This particular feature is called the tweet update speed and is unique only to Twitter. In terms of interaction, this constant and sustained rate of communication gives Twitter a clear edge over competing social media tools and even traditional modes of news delivery.

2) Twitter's prowess as an adept news breaking tool: Three incidents demonstrate Twitter's capability to deliver up-to-date breaking news at warp speed even in the absence of the media. Within minutes of the November 26, 2008, Mumbai, India terrorist attacks, Twitter was providing an up-to-date stream of information through the searchable tag #mumbai. This was well before any TV station or newspaper had broken the news. Britain's daily The Guardian describes this incident as the moment Twitter came of age as a vital news dissemination service (Arthur, 2008, November 27; Lewis, 2008, November 28).

However, it is a March 5, 2009, explosion in downtown Bozeman, Montana that provided the "aha" moment that propelled Twitter's role as a legitimate news-gathering and distribution tool. With no journalists on site, early reports of the explosion, which destroyed half a block while sending debris flying 200 feet, were mostly available only on personal Twitter updates. These updates included photos, descriptions from the scene and emergency information. Soon after, several media organizations were reporting based on these tweets while directing their readers and listeners to the same tweets for updates (Lowery, 2009).

Recently Twitter played a similar role in the Arab Spring, first as a news coverage tool then as a mobilization tool. Social media maven and NPR's strategist Andy Carvin embodies the former. Starting with the Tunisian uprising in late last 2011, Carvin continually posted news updates from his Twitter account (@acarvin) on the Egyptian and Libyan uprisings (Katz, 2011, March 14). With the increased presence of mainstream media outlets on Twitter, the likelihood of the Mumbai and Bozeman scenarios playing out in a similar fashion is low. Most media outlets now have some form of presence on Twitter, ranging from USA Today (@USATODAY) to The Alaska Anchorage Daily (@adndotcom).

3) Dwindling news readership: I discuss both this factor and the increasing public distrust of media agencies in later sections of this paper. On a positive note however, these trends have been accompanied by an uptick in online news readership. Also, traditional blogs have continued to enjoy better credibility ratings than mainstream media agencies (Johnson, Kaye, Bichard & Wong, 2008; Johnson & Kaye, 2010). The above mentioned factors add impetus to the role SMEs play in contemporary media practice. It means that as the faces of the media agencies in the digital world, they are now best positioned to control how news flows into the social media sphere, how the media agencies interact with audiences in the digital world and eventually bring them into the fold. It is important to study how these SMEs interact with audiences and what framing patterns emerge in the news articles they post in the social media sphere.

Because the social media editors represent different media formats, say TV and print news, it is also wise to examine if format plays a role in terms of interaction and news framing. Hence, the purposes of this paper are: a) to determine if interaction, measured by the personalization of tweets, varies according to the media format an SME represents, namely print news and TV; b) to use media framing theory as a guide to examine the news frames within the articles these editors linked to their Twitter posts; and c) to determine if these framing patterns varied according to the media format the SMEs represent.

Defining Twitter

Twitter describes itself as "a real-time information network that connects [users] to the latest stories, ideas, opinions and news about what [they] find interesting" (Twitter.com, 2013, May 5). Having been in existence for about seven years, Twitter has emerged as the most popular microblogging site with estimates of anything between 200 to over 500 million active users (Chapman, 2011, March 21; Pan, 2012, February 22), most of who send about 400 million tweets daily (Tsukayama, 2013, March 21). Users post status updates called "tweets" on their profile pages, which are in turn read by their "followers" (see Table 1). With only 140 characters allowed per tweet, users have to send short, pithy bursts of information that can also contain links to stories, images and videos.

Personalization, a facet of this study, plays a crucial role in social media communication and more so on Twitter. Twitter allows an individual to build a set of dedicated and committed followers who basically make up what would otherwise be called an entourage. This means that when someone signs up to follow a particular SME, chances are that there is a level of personal interest in that SME as an individual rather than as a mere extension of a particular news agency. This study seeks to find out how the SMEs reciprocate to these overtures from their Twitter followers through personalization.

Table 1: Definition of Twitter Terms

Retweet:

An original post repeated and forwarded by another user in order to propagate news. Retweets are commonly identified as such in a message or with the abbreviation RT.

Reply:

A tweet posted in response to another tweet. A reply is usually identified by the "at symbol", @.

Mentions:

These are very similar to replies and occur when a user posts another user's username within their tweet. Mentions are also identified by the "at symbol", @.

Keywords:

These are identified by the hash tag (#). Users insert the hash tag before a keyword or topic. When clicked, the keyword brings up all tweets that are relevant to that keyword or topic.

Plain:

These are tweets that contain no replies, retweets or mentions.

Via:

Twitter indicates where a tweet was posted from. Common sources include; Twitter, TweekDeck, various mobile devices and the Web.

Followers:

Not all Twitter users post status updates. Some simply follow those who do by subscribing to their tweets. In turn, they receive status updates on their accounts. Both statistics appear on a user's profile page.

Blogs and Media Framing

In a broad sense, framing refers to the selection of certain aspects of reality in order to make them more salient in a bid to promote a desired interpretation (Entman, 1993). Narrowly defined, media frames refer to the sense-making and interpretive packages journalists use to contextualize events by manipulating metaphors, catchphrases and images in order to prioritize some events over others (Gamson & Modigliani, 1989; Pippa, Montague & Marion, 2000). Media frames generally fall into two categories: generic and issue-specific frames. Generic frames are broad and structural themes and are limited to conflict, human interest, economic impact, responsibility and morality (Semetko & Valkenburg, 2000).Issue specific frames are more subjective. They are flexible and vary depending on the content being analyzed and they change based on the topic under study and the prevailing context (de Vreese, 2005). This study examines both the generic and issue-specific frames in the news articles posted by social media editors via Twitter.

Journalists traditionally use the conflict frame to portray adversarial dynamics between individuals, groups or institutions. They use the human interest frame to emotionalize issues and news events. Similarly, they employ the economic impact frame on news coverage to portray the impact thereof on individual players, institutions or nations. The responsibility frame connects certain events and the accompanying consequences or solutions while the morality frame emphasizes religious and morality issues in news coverage (Semetko & Valkenburg, 2000).

While media framing is commonly used to examine news coverage, there is a dearth of scholarly research as pertaining to framing and microblogging. A possible reason could be the newness of this medium, namely Twitter, and recently Tumblr. A recent study found that via Twitter, and with regional variation, most media agencies emphasize crime, public affairs, lifestyle and business news topics the most (Armstrong & Gao, 2010). While the above mentioned study did not specifically use the framing theory, its findings still indicate that there exists some similarity between Twitter and traditional framing patterns. It is important to point out that the Armstrong and Gao study only examined the content of the tweets and not the linked stories. Additionally, the study looked at the general/corporate Twitter accounts maintained by the different agencies and not necessarily those run specifically by the SMEs.

Regardless, the better portion of pertinent research has focused on framing and traditional blogs. Blogs, also known as Web logs, first emerged as online journals where people could post personal musings and thoughts. It was only after September 11, 2001, and the subsequent rise of war bloggers that the blogosphere morphed into an arena of political, academic and journalistic punditry (Gallagher, 2002, June 10). Soon after, researchers started examining the relationship between blogs and the news media (Goldman & Kuypers, 2010).

Guillory (2007) found that science and technology blogs emphasized the conflict frame, regardless of the political leanings of the blogger. These bloggers also downplayed the economic impact frame. In a multidimensional frame analysis study, Bichard (2006) examined the blogs run by candidates George W. Bush and John Kerry during the 2004 presidential elections. Both candidates equally emphasized the time frame, an issue-specific frame, by focusing their posts in the context of the present time. The candidates, however, differed in the way they presented the space dimension frame to their supporters. John Kerry was more likely to emphasize the societal and international frames while Bush emphasized the individual and community frames. While the blog posts on both candidates' sites tended to have a neutral tone, Kerry's blogs carried a majority of the negative posts appearing on both sites.

Similar research shows that the framing of political blogs affects participation levels in political campaigns. In a study of campaign mobilization, Thorson and Wise (2007) found that those blogs that emphasized campaign strategy in their mobilization requests were more successful than those that emphasized issue-related frames such as a candidate's stand on different social issues. Mobilization strategy frames included those that focused on winning and losing, poll results and political tactics.

Blogs can also mobilize citizen action outside of electoral situations. Such evidence emerged during a contentious debate on pay raises for Louisiana lawmakers during the summer 2008, economic crisis. User comments posted on a popular TV station's blog (WAFB-TV of Baton Rouge) contained certain thematic frames that also had an agenda setting effect on the channel's news coverage. The three thematic frames that emerged were: living conditions, political corruption and Gov. Bobby Jindal's performance. While initial TV news reports focused on a more balanced approach, later reports reflected the blog's themes: public suffering, political corruption and Jindal's failure to address these issues. Needless to say, the governor vetoed the bill (Biswas, 2009).

While the Louisiana pay raise debate demonstrated that bloggers and mainstream news framing may converge at some point, this is not always the norm. Anderson (2010, June) found that blogs and mainstream newspapers framed the 2007 congressional shield law debate differently. Shield laws usually protect journalists from forced disclosure of sources and subsequent contempt of court charges (Trager, Russomanno & Ross, 2007; Zelezny, 2007).While covering the same debates, bloggers emphasized political and legislative developments more than the newspapers did. Since shield laws have to do with journalistic privileges and protections from subpoenas and similar legal actions (Overbeck, 2007), a big question in the debate was who qualifies to be a journalist? With their First Amendment rights at stake, bloggers were keen to emphasize this part of the debate more so than newspapers. Hence, they framed a more generalized and wider definition of who a journalist is and the accompanying privileges and protections. They also did this in more positive terms than the newspapers.

Media Format and News Coverage

This study partly examines whether news framing and media format (TV vs. print news) are related. Indeed, research shows that such a relationship exists, not only in news framing but also in terms of journalistic values, which in turn might influence news framing patterns. Since no studies have looked specifically at professional and news values vis-à-vis media format as pertaining to social media, a close comparison could be the famed Weaver and Wilhoit studies on journalism's professional values. Their 2000 study on print, broadcast and online journalists' values found differences in the manner these journalists viewed their roles in terms of the three journalistic functions: disseminator, interpretive and adversarial. Within the disseminator function, broadcast journalists were most likely to avoid unverifiable facts and strived more to reach a wider audience. Within the interpretive function, online journalists were more likely to investigate government claims, analyze complex problems and discus both national and international policy. Both online and print journalists were more likely to take an adversarial role against government officials and businesses than broadcast journalists (Weaver et al., 2007).

Other research shows similar divergence based on media format. In a study of online and print journalists, Brill (2000) found stark differences in the interpretive function. A majority of traditional journalists (90%) viewed this function to be important while only 24% of online journalists concurred. The study also found that online journalists favored their role as interpreters of complex issues more than print journalists by a margin of 19 percentage points. However, print journalists favored giving information to the public more and also were more adversarial towards the government and business. While journalistic roles and professional values might not be the sole determinants of news framing, they however tie into news framing and media format.

Studies similar to those discussed above have also shown a relationship between media format and news framing. This is true even in this age of convergence, where different media platforms share content and lines between formats are blurred. In a study of the visual coverage of the Iraqi invasion, Keith, Schwalbe, and Silcock (2009) found distinct differences in the types of images published across different media formats. Even though TV and print news emphasized images of the U.S. war machine at the onset of the war, TV news broadcast significantly more images of journalists than did print and online news. Additionally, this trend sustained through week five of the invasion. Although the above mentioned study did not specifically look at news framing, the portrayal journalists in a war situation could depict the human interest frame.

Similarly, bloggers and online news practitioners frame their news differently. Kyounghee and Shin-Il (2011) examined how news outlets and blogs framed the 2008 Virginia Tech massacre. Blogs were more likely to frame news at the societal level while newspapers largely balanced their coverage at the individual, community and societal levels. These framing dynamics are not limited to the American media. Semetko and Valkenburg (2000) found similar framing trends in the Dutch media coverage of the 1997 "Eurotop" summit of the European Union leaders. Dutch newspapers were most likely to emphasize the responsibility, conflict and economic frames. TV stations broadcast more human interest and morality frames.

Hypothesis 1: There is a difference regarding the occurrence of generic frames between articles posted by TV SMEs and those posted by print news SMEs.

This hypothesis predicts that articles posted by SMEs from print media carried different generic frames from those posted by TV news SMEs. This hypothesis therefore predicts a relationship between media format and generic news framing.

Hypothesis 2: There is a difference regarding the occurrence of issue-specific frames between articles posted by TV SMEs and those posted by print news SMEs.

Similarly, this hypothesis predicts that SMEs from print media and TV news posted articles that differed in their emphasis on issue-specific frames (see Table 2).

Personalization and Audience Interaction

Mayer (2011) warns of the danger that arises when journalists distance themselves from their audiences in the pursuit of objectivity and fairness in news reporting. She hearkens back to the nineties, when then emergent public journalism faced criticism for pandering to audiences. With the advent of online journalism, such criticism would seem misplaced, and more so given the interactive nature of social media. Mayer further states that journalists may seek interaction with audiences in order to get more information about their audiences' needs and to make the audience feel more connected to the news agency. This is what SMEs seek. Readers seem to reciprocate this.

Chung (2009) found that readers expected journalists to adopt the populist mobilizer and public journalism functions. The mobilizer role is part of the Weaver and Wilhoit journalistic functions discussed earlier. Within the mobilizer role, Chung's study showed that readers expected journalists to motivate ordinary people and give them a chance for self-expression while pointing people towards solutions. Readers expected reporters to investigate citizen priorities, motivate people and include them as sources under the civic journalism role. While an SME's Twitter followers might not necessarily have direct input to editorial material, the interaction therein brings the news agency closer to those it serves. This takes a premium given that readership rates have fallen over time while trust in the media has also declined similarly.

Overall, news readership in the U.S. has dropped despite the fact that more people access news online. The Pew Research Center (2009, February 26) reports a four point drop in overall readership between 2006 (43%) and 2008 (39%). Newspaper readership in particular has fared worse since then and now stands at 31%. Print news readership has a stand-alone readership of 25%. Meanwhile, the public's trust in the media in terms of accuracy and impartiality has also fallen. Sixty-three percent of Americans believe that news organizations produce inaccurate stories while 60% state that news organizations are politically biased (Pew Research Center, 2009, September 12). It is with this backdrop of distrust and cynicism that news organizations need to revitalize erstwhile robust relationships with their audiences. The highly interactive nature of Twitter might go a long way in allowing SMEs to do just that, by personalizing their tweets even if it is only with the readers who follow them on Twitter.

Hypothesis 3: There is a difference regarding how TV and print SMEs personalize their tweets

This hypothesis predicts that there was a difference in how SMEs from print media and TV news personalized their tweets. Personalization was measured by the extent to which a tweet emphasized the editor's personal life such as a hobbies, leisurely activity, humor or similar non-news related information.

Research Questions

RQ1: What were the most common generic frames in the articles linked via Twitter?

This research question seeks to find which of the five generic frames occurred with the highest frequency within the articles posted by the social media editors via Twitter; conflict, human interest, economic impact, responsibility and morality. A sixth generic frame--technology--was included after an informal survey of the tweets showed that some articles were exclusively technology related.

RQ2: What were the most common issue-specific frames in the articles linked via Twitter?

de Vreese (2005) advises that because issue-specific topics vary depending on the topic of study and prevailing circumstances, using a priori categories while simultaneously avoiding speculative content analysis methods in favor of more deductive methods is more appropriate. Furthermore, these framing categories must be: easy to discern, be common in journalistic practice, be unique from other frames and be cognizant to others outside of the purview of the interested parties. Hence, this study used topics from the 2007 Pew Research Center report on the most followed news topics in the U.S. (Robinson, 2007, August 15) to examine RQ2. These are listed in Table 2.

Table 2: Issue-Specific Frames

War/Terrorism

Man-Made Disasters

Natural Disasters

Business

Crime and Violence

Health and Safety

Politics

Sports

Science and Technology

Lifestyle

Entertainment

Religion

Oddity

Other

Method

This study used content analysis based on three sets of four weekdays (Monday - Thursday) for a total of twelve days between April 12 and May 2, 2011. Fridays and weekends were excluded after the pilot study indicated that a majority of the tweets posted on those days tended to be purely personal and non-news related.

A random sample of eight SMEs was derived from the nation's top newspapers: USA Today, Wall Street Journal, Washington Post and the LA Times; and four major TV stations: ABC News, NBC, CNN, and MSNBC. A comprehensive list of SMEs is published and maintained on Twitter by Sree Sreenivasan of Columbia University.

The unit of analysis was a single tweet and coders analyzed the ten most recent tweets for each editor on a given day. Coders first read the tweet and determined how personalized it was, and then clicked on any hyperlinks within the tweet and coded the linked articles. Using skills from prior training, coders then determined if the linked articles fell under the six generic frames or the 14 issue-specific frames. In order to account for time zone differences, coding was done no earlier than 11a.m. CST and no later than 3 pm CST. The inter-coder reliability scores for the three categories on which the coders had to make subjective evaluations were: Generic frames (pi = .87); issue-specific frames (pi = .86) and personalized tweets (pi = .91).

It is important to point out that while this study does not necessarily imply that the SMEs authored the articles analyzed, it is reasonable to assume that they had a say as to which type of articles they linked to their Twitter feeds. Additionally, the Twitter feeds analyzed were specifically maintained by the said SMEs under their identity as representatives of a particular news agency. The sample does not include the general/corporate Twitter accounts maintained by any one news agency.

Operationalization of Coding Categories and Definition of Twitter Terminology

The following is the operationalization of the coding categories for the generic frames and personalization of the articles linked to the SMEs' tweets:

Conflict: An article qualifies if it emphasizes conflict or disagreement. For example, political wrangling or war activities qualified as conflict.

Human interest: Articles that emphasize demotion and human life stories fell under this category. Examples would include: humanitarian work, rescue operations after a disaster or the biography of a famous person.

Economic consequence: This frame encompasses articles that emphasized economic issues and trends, such as corporate quarterly earnings, corporate takeovers, or the state of the U.S. economy.

Morality frame: Articles that fell under this frame dealt with moral tenets and issues of social prescription. Examples would be political scandal, fraudulent activity or celebrity scandals.

Responsibility frame: Articles related to this frame describe certain events, problems or issues in a manner to attribute responsibility or solutions.

Technology: These articles focused strictly on technology based news such as new social media applications, gadgets, etc. Examples would be new apps, new web-based services, interactive technology, software upgrades or social media news.

Personalization: This refers only to the tweets, and not the stories linked to them. Personalized tweets represented an SME's personal life, humor, leisure activities or in-transit experiences.

Mainstream source: Stories that originated from well know media agencies like CNN, The New York Times, NPR, and The Telegraph were coded as mainstream sources.

Breaking news: A linked article qualified as breaking news if the timeline indicated that it was published within the last hour.

Results

A total of 950 tweets were analyzed over the coding period. Most tweets posted were plain (31%), meaning that they were not replies, retweets or did not contain any mentions. Replies were the second most common tweets and made up 30% of all tweets posted. Retweets accounted for 19% all tweets posted while 21% of all tweets posted were based on a mention. Most SMEs (39%) preferred to post their tweets via TweetDeck, a mashup service while 21% posted via unspecified Web services. Only 6% posted via handheld devices and even fewer (1%) posted directly from Twitter. Slightly more than half of all tweets posted were personal and about 46% also included a link. Of all linked articles, a majority were news stories (93%). Of these, 86% were from mainstream sources. Twenty-six percent of all linked news stories qualified as breaking news.

Hypothesis 1: There is a difference regarding the occurrence of generic frames between articles posted by TV SMEs and those posted by print news SMEs

Hypothesis 1 postulated that there is relationship between media format and generic news framing. The hypothesis predicted differences in the frequency that generic frames would occur between articles posted by SMEs from either media format. Chi-square analysis shows that indeed such a relationship exists. Table 3 indicates that while TV SMEs were more likely to emphasize technology in their linked articles, print news SMEs favored the rest of the generic frames. In order to preserve the validity of the chi-square results, categories that scored less than five were excluded in the analysis, namely morality and responsibility.

Table 3: Chi-Square Test for Generic Frames

Generic Frames

Print

TV

Total

Technology

Human Interest

Conflict

Economic

88

85

36

23

96

69

24

8

184

154

60

31

?2 = 8.872 | df = 3 | p = 0.031 | N = 429

Hypothesis 2: There is a difference regarding the occurrence of issue-specific frames between articles posted by TV SMEs and those posted by print news SMEs

Chi square analysis also supported Hypothesis 2 (Table 4). There is a significant relationship between media format and the occurrence of issue-specific frames in the articles linked via the SMEs' tweets. Print media SMEs were most likely to post articles that emphasized science & technology alongside war & terrorism. TV SMEs linked more entertainment related articles. By sizeable margins however, print news SMEs posted more articles that emphasized politics, lifestyle and business frames. All SMEs largely posted few links to articles related to the rest of the issue-specific frames. For this reason, and to preserve the validity of the chi-square results, the following categories were excluded from the analysis: crime and violence, man-made disasters, health and safety, religion and oddity.

Table 4: Chi-Square Test for Issue-Specific Frames

Generic Frames

Print

TV

Total

Sciences & technology

War & Terrorism

Entertainment

Politics

Lifestyle

Business

Sports

Natural disasters

Other

81

36

23

27

17

11

5

6

25

78

27

36

18

8

2

4

6

13

159

63

59

45

25

13

9

12

38

?2 = 16.07 | df = 8 | p = 0.041 | N = 423

Hypothesis 3: There is a difference regarding how TV and print SMEs personalize their tweets

This hypothesis sought to find out whether there were any differences in the extent to which the SMEs personalized their tweets across media format; print media and TV news. Chi-square analysis indicates that indeed there is a relationship between personalization and media format. Table 5 shows that TV news SMEs were significantly more likely to personalize their tweets.

Table 5: Chi-Square Test for Personalization of Tweets

Generic Frames

Print

TV

Total

Personalized Tweets

Non personalized Tweets

218

248

263

212

481

460

?2 = 6.942 | df = 1 | p = 0.008 | N = 941

RQ1: What were the most common generic frames in the articles linked via Twitter?

Technology (43%) was the most common generic frame in the articles linked via the SMEs' tweets (Table 6). Human interest was the second most common generic frame accounting for 36% of all articles linked. Conflict came in third while economic consequence and morality were the least common frames.

Table 6: The Most Common Generic Frames for Linked Articles

Generic Frames

Print

TV

%

Technology

Human Interest

Conflict

Economic

Morality

88

85

36

23

0

96

69

24

8

2

43.00

36.00

14.00

07.00

00.04

N = 431

RQ2: What were the most common issue-specific frames in the articles linked via Twitter?

Science & technology (36%) was by far the most common issue-specific frame among all articles linked from Twitter (Table 7). War & terrorism and entertainment accounted for approximately 14% of all issue specific frames among all linked articles. While politics and lifestyle frames scored modestly at about 10 and six percent respectively, all other issue-specific frames registered a miniscule presence.

Table 7: The Most Common Issue-specific Frames for Linked Articles

Generic Frames

Print

TV

%

Sciences & technology

War & Terrorism

Entertainment

Politics

Lifestyle

Business

Sports

Natural disasters

Health & safety

Crime & violence

Man-made disasters

Oddity

Religion

Other

81

36

23

27

17

11

5

6

2

2

1

2

1

25

78

27

36

18

8

2

4

6

1

2

1

2

1

13

36.30

14.40

13.50

10.03

05.70

03.00

02.10

02.70

00.07

00.09

00.05

00.09

00.05

08.70

N = 438

Discussion

This study was designed to examine how social media editors use their Twitter accounts to interact with audiences and how they shape and frame the news via the articles they link to their tweets, and whether these trends vary depending on the media format. Chi-square analysis supported hypotheses one and two regarding news framing and differences per media format. Print media and TV news SMEs posted articles that carried different frames. TV SMEs emphasized the technology frame the most, by five percentage points over print news SMEs. By a large margin (11%), however, print news SMEs favored human interest stories. They also outscored TV SMEs on the conflict and economic impact frames. This shows that while there are commonalities in the frames SMEs choose to emphasize overall, there is variation in how they do it. Some are more vigorous in pushing certain frames while others, who might not necessarily discount those particular frames, will choose to emphasize others.

The findings also supported Hypothesis 3, which sought to find out if and how the SMEs personalized their tweets. Overall more of the tweets posted were personal than non-personal. However, TV SMEs were more likely to personalize their tweets than their print news counterparts. Personalization of tweets should not come as a surprise since SMEs primarily serve as the newsroom's liaison to the digital world. They are the agency's face in the social mediasphere. It is they who interact with readers on a micro level and, therefore, it is prudent that they personalize their tweets in order to imbue the agency-client relationship with a human touch.

As stated in the literature review, Twitter avails itself to personalization since it allows an individual to build a set of dedicated and committed followers who have a personal interest in a particular SME. In turn, the SMEs in this study reciprocated and cemented this relationship by tweeting things about their personal life, humor, in-transit activities, etc. Additionally, most SMEs went ahead and specifically emphasized personal themes in their Twitter bio blurbs. Apart from other personal information such as native cities, spouses and hobbies, it was common to encounter phrases such as; music junkie, foodie, skier, surfer (badly), dad, gummi worms (sic), etc.

Research questions 1 and 2 show that technology-related frames trumped all else within the articles posted via Twitter. All SMEs emphasized this generic frame the most, outscoring human interest by seven percentage points. By a large margin, science & technology was the most common issue-specific frame. Science & technology outscored war & terrorism, the next most common issue-specific frame by 14 percentage points. Three reasons can explain why the emphasis on technology should come as no surprise.

In general, most SMEs are young, tech savvy, and geared towards the tech world. The SMEs examined in this study were no different. Also, the manner through which they updated their tweets is testament. Most SMEs (39%) used TweetDeck to update their accounts. In fact only 10 out of the 950 tweets examined in this study where posted directly through Twitter. TweetDeck is a popular mashup website that allows users to manage several social media accounts within one convenient interface. Users can update their Facebook, MySpace, LinkedIn, Twitter and Foursquare accounts within a single browser. TweetDeck is also available for hand held devices such as Androids, iPads, and iPhones. It would be safe to assume that at any one time an SME will be dealing with audiences from several social media sites. The fact that SMEs have to explore all facets of the social media sphere through mashup sites might also explain why they tended to emphasize technology in their tweets. Moreover, a telling clue would be the various monikers some have adopted to describe their positions, eschewing the normative social media editor title. Coders came across such terms as; digital journalist, digital audience developer, digital media producer, digital editor, etc.

The human interest frame was the second most common generic frame while conflict was a distance third, trailing by thirty-five percentage points. This shows that SMEs prefer technology and human interest stories overwhelmingly over conflict-related news. Ironically, conflict happens to be a staple in regular news coverage (Semetko & Valkenburg, 2000). Even though war and terrorism was the second most common issue-specific frame - and its being a conflict related frame - does not counter the above argument. Science and technology still outscored it by fourteen percentage points. Furthermore, war & terrorism outdid entertainment only by a single percentage point.

The de-emphasis of conflict and the favorability towards human interest and technology highlights an interesting trend in how SMEs selected the news articles they posted via Twitter. Research shows that science and technology, as well as entertainment are among the least followed news categories (Robinson, 2007, August 15). However, this study shows that the SMEs emphasized these categories. Disaster, money, conflict and political related news are the most followed news categories (Ibid). Additionally, the aforementioned Armstrong and Gao (2010) study on the most popular topics tweeted by news agencies had crime in the lead. Public affairs came in second and business news fourth. My findings indicate that the SMEs downplayed the crime and disaster frames while the economic frame accounted for a miserly 7 % of all generic frames. More so, politics came in fourth among the issue-specific frames.

Implications for Media Agencies and Audiences

While this study is by no means a definitive and comprehensive portrayal of social media editors and how they frame the news, or even how they determine to do so, the findings provide two interesting insights: 1) Are SMEs framing the news counter to the dictates of the editorial office, and if so, then 2) Are they providing news readers the type of news they prefer? As discussed earlier, there clearly is a disjuncture between what types of news frames the mainstream media normally emphasizes (even via Twitter) and what the SME's are providing via Twitter. One possible explanation to this disjuncture would be that the SMEs are demographically young and thus drawn to tech-oriented frames. Secondly, it might be that those who follow them via Twitter are equally young and tech savvy and largely consume tech-oriented media. This might mean that those who follow the SMEs on Twitter do so because they find commonality with them. This proposition would then make sense if we assume that the SMEs are tailoring their frames to cater to a niche, tech-oriented audience.

Thirdly, it might be that the SMEs are a group of tech-focused journalists who are trying to find a balance in the rapidly changing and demanding world of social media. This means that while they might emphasize technology in their tweets, they also find room for other popular news. This might explain why human interest was the second most common generic frame, and war and terrorism the second most common issue-specific frame. Regardless, the SMEs' main purpose, which is to increase traffic to the parent site and thus increase readership would still be served under the three scenarios mentioned above. Given that news readership rates have fallen drastically lately, it forces news agencies to lure readers and viewers from all demographics and by whichever methods they can deploy. One way to do that is to delve into the digital world and through interaction and the emphasis of news frames that are compatible with the denizens of that world, attract more readers while quenching their thirst for niche news.

Combining the above mentioned thoughts to the novelty of the SME position and the fluidity of the world in which these editors operate means more research is appropriate. Probable questions could probe: Whether framing varies according to the size of an agency?; Are there any geospatial factors that come into play?; Do opinionated news agencies differ?; How much control does the editorial office exercise (if at all) in determining the types of articles posted via Twitter?; and lastly, are there any unique news values these SMEs might hold?

This last query would probably be the most useful in developing a comprehensive profile of the SMEs. Future research might need to look into the journalistic practices and values these SMEs hold. Specific guidelines would include: demographics, working conditions, perceptions, political orientation, adaptability to technological changes, ethics, and lastly news values. Such research could be fashioned after the renowned Weaver and Wilhoit's longitudinal studies of the state of American journalism (Weaver & Wilhoit, 1991; 1996; Weaver et al., 2007). Not only would these findings delineate the specific challenges that SMEs face in the social media sphere, they might also provide insight as to why and how SMEs frame their news counter to the norm.

References

Anderson, Christopher W. (2010, June). A comparative content analysis of newspaper and weblog reporting on attempts to pass a media shield law. Paper presented at the annual meeting of the International Communication Association, Singapore.

Armstrong, Cory L. & Gao, Fangfang. (2010). Now tweet this: How news organizations use Twitter. Electronic News, 4(4), 218-235.

Arthur, Charles. (2008, November 27). How Twitter and Flickr recorded the Mumbai terror attacks. The Guardian. Retrieved November 27, 2008, from http://www.guardian.co.uk/technology/2008/nov/27/mumbaiterror-

attacks-twitter-flickr.

Bichard, Shannon L. (2006). Building blogs: A multi-dimensional analysis of the distribution of frames on the 2004 presidential candidate web sites. Journalism & Mass Communication Quarterly, 83(2), 329-345.

Biswas, Masudul. (2009). Discourse analysis of online discussion and news stories on the legislative pay raise bill in Louisiana. Journal of New Communications Research, 4(1), 47-65.

Brill, Ann M. (2001). Online journalists embrace new marketing function. Newspaper Research Journal, 22(2), 28-40.

Chapman, Glenn (2011, March 21). Twitter marks fifth birthday. AFP via Yahoo News. Retrieved March 21, 2011, from http://news.yahoo.com/s/afp/20110321/tc_afp/

usitinternetcompanytwitter_20110321174406fpew.

Chung, Deborah S. (2009). How readers perceive journalists' functions at online community newspapers. Newspaper Research Journal, 30(1), 72-80.

deVreese, Claes H. (2005). News framing: Theory and typology. Information Design Journal & Document Design, 13(1), 51-62.

Entman, Robert M. (1993). Framing: Towards clarification of a fractured paradigm. Journal of Communication, 43(4) 51-58.

Gallagher, David F. (2002, June 10). A rift among bloggers. The New York Times. Retrieved June 1, 2013, from http://www.nytimes.com/2002/06/10/business/technology

-a-rift-among-bloggers.html?src=pm.

Gamson, William A. & Modigliani, Andre (1989). Media discourse and public opinion on nuclear power: A constructionist approach. American Journal of Sociology, 95, 1-37.

Gleason, Stephanie. (2010). Harnessing social media. American Journalism Review, 32(1), 6-7.

Goldman, Adria Y. & Kuypers, Jim A. (2010). Contrasts in news coverage. Relevant Rhetoric: A New Journal of Rhetorical Studies, 1-18.

Guillory, Brant. (2007). A framing analysis of science and technology weblogs: How is science presented by commentators? Paper presented at the annual meeting of the International Communication Association, San Francisco.

Hazel, Kwon Kyounghee & Moon, Shin-Il. (2009). The bad guy is one of us: Framing comparison between the us and Korean newspapers and blogs about the Virginia Tech shooting. Asian Journal of Communication, 19(3), 270-288.

Katz, Ian. (2011, March 14). SXSW 2011: Andy Carvin--the man who tweeted the revolution. The Guardian. Retrieved June 1, 2013, from http://www.guardian.co.uk/technology/2011/mar/14/andy-carvin-

tunisia-libya-egypt-sxsw-2011.

Keith, Susan, Schwalbe, Carol B. & Silcock, William B. (2009). Visualizing cross-media coverage: Picturing war across platforms during the U.S.-led invasion of Iraq. Atlantic Journal of Communication, 17(1), 1-18.

Lewis, Paul. (2008, November 28). Twitter comes of age with fast reports from the ground. The Guardian. Retrieved November 28, 2008, from http://www.guardian.co.uk/world/2008/nov/28/mumbai-terrorattacks

-india-internet-technology-twitter.

Lowery, Courtney. (2009). An explosion prompts rethinking of Twitter and Facebook. Nieman Reports, 63(3), 32-34.

Mayer, Joy. (2011). Engaging communities: Content and conversation. Nieman Reports, 65(2), 12-14.

Overbeck, Wayne. (2007). Major principles of media law. Belmont, CA: Thomson Wadsworth.

Pan, Joann. (2012, February 22). Will you be Twitter's 500 millionth user? Mashable.com. Retrieved June 1, 2013, from http://mashable.com/2012/02/22/twitters-500-million-user.

Pew Research Center. (2009, February 26). Newspapers face a challenging calculus. Pew Research Publications. Retrieved June 1, 2013, from http://pewresearch.org/pubs/1133/decline-print-newspapers

-increased-on.

Pew Research Center. (2009, September 12). Press accuracy rating hits two decade low: Public evaluations of the news media: 1985-2009. Retrieved June 1, 2013, from http://www.people-press.org/2009/09/13/press-accuracy-rating-hits-

two-decade-low.

Pippa, Norris, Montague, Kern & Marion, Just. (2000). Framing terrorism: The news media, the government and the public. New York: Routledge.

Robinson, Michael J. (2007, August 15). Two Decades of American news preferences. part 1: Analyzing what news the public follows --- and doesn't follow. Pew Research Center for the People and the Press. Retrieved June 1, 2013, from http://pewresearch.org/pubs/566/two-decades-of-american

-news-preferences.

Semetko, Holli & Valkenburg, Patti. (2000). Framing European politics: A content analysis of press and television news. Journal of Communication, 50(2), 93-109.

Thorson, Kjerstin & Wise, Kevin. (2007). Blogging for participants: Framing the candidate blog for mobilization. Paper presented at the annual meeting of the International Communication Association, San Francisco.

Trager, Robert, Russomanno, Joseph & Ross, Susan D. (2007). The law of journalism and mass communication. New York: McGraw-Hill.

Tsukayama, Hayley. (2013, March 21). Twitter turns 7: Users send over 400 million tweets per day. The Washington Post. Retrieved June 1, 2013, from http://articles.washingtonpost.com/2013-03-21/business/37889387

_1_tweets-jack-dorsey-twitter.

Twitter.com. (2013, May 5). The fastest, simplest way to stay close to everything you care about. Retrieved June 1, 2013, from https://twitter.com/about.

Weaver, David H. & Wilhoit, Cleveland G. (1991). The American journalist: A Portrait of U.S. news people and their work. Mawhaw, N.J.: Lawrence Erlbaum Associates.

Weaver, David H. & Wilhoit, Cleveland G. (1996). The American journalist in the 1990s: U.S. news people at the end of an era. Mawhaw, N.J.: Lawrence Erlbaum Associates.

Weaver, David H., Beam, Randal A., Brownlee, Bonnie J., Voakes, Paul S. & Wilhoit, Cleveland C. (2007). The American journalist in the 21st century: U.S. news people at the dawn of a new millennium. Mawhaw, N.J.: Lawrence Erlbaum Associates.

Zelezny, John D. (2007). Communications law: Liberties, restraints, and the modern media. Belmont, CA: Thomson Wadsworth.

About the Author

Ben S. Wasike, Ph.D., is an Associate Professor at the University of Texas at Brownsville, United States. His research interests include new media, social media and the convergence of social media, and political communication.

Citing this paper:

Wasike, Ben S. (2013). Framing news in 140 characters: How social media editors frame the news and interact with audiences via Twitter. Global Media Journal -- Canadian Edition, 6(1), 5-23.