BREAKING: #J OURNALISM @T WITTER

(1)

BREAKING:

#J

OURNALISM

@T

WITTER

A

STUDY ON HOW JOURNALISTS AT

RTL

AND

NOS

USE TWITTER

Esther de Kloe

Student number: 1912089

Master Thesis Journalism Studies

Rijksuniversiteit Groningen

August 2011

(2)

2

Acknowledgements

This thesis is the closing piece of my MA Journalism Studies at the Rijksuniversiteit Groningen. It reflects the ending of two of the best, but also the hardest years of my time in university.

I want to thank my supervisor Todd Graham for his constructive comments and for guiding me trough the writing process. I am grateful to my friends, family and especially my father. They always kept believing in me and I could always count on them for support.

This thesis is dedicated to my mother. I wished that she could have seen me graduate. But even though she will not be physically there, she always is with me in my thoughts.

Esther de Kloe

(3)

3

Introduction

‘The best way to discover what’s new in your world.’(Twitter.com, 2010). This was the slogan of micro-blogging and social network site Twitter.com. In messages of only 140 characters, Tweeps1 can answer the question: “What’s happening?” They also have the possibility to add a picture or hyperlinks to their message. People who choose to follow a person will see his or her tweets2 automatically. That people are interested in what is happening and what other people are tweeting is shown by the rapid growth of the amount of active users. In the year 2009, the number of users increased 900 percent (Mirck, 2009). In June of that year, there were 45 million unique users of Twitter (Schonfeld, 2009). And in July of this year Twitter had 200 million users (Qazi, 2011). How many of these users are Dutch Tweeps is unknown. Also the amount of posted tweets show that Twitter is getting more and more popular. On the first of August in 2010 the 20 billionth tweet was posted. It took 4 years for the first 10 billion tweets to be posted. But only five months this number was doubled to 20 billion (Batty, 2010). And In June 2011 a billion tweets are sent every five days (Twitter Blog, 2011).

Twitter could be a useful tool for journalists, but the number of journalists using Twitter in the Netherlands is not clear. In an article about the discovery of Twitter by journalists, Verbraeken (2009) wrote that there are 13.000 Tweeps in Holland of whom 500 are journalists. In that same article, Verbraeken poses that there are 4 to 5 million Tweeps worldwide (Verbraeken, 2009). Since she wrote her article in February so she could be right, as we have seen Twitter grow rapidly in 2009. The Dutch Twitter account JournalistenNL follows 584 people (August, 2011), who claim to be a Dutch journalist.

News organizations have been quick to adopt Twitter (Fahri, 2009). Journalists tend to adopt new media as they develop. Much research has been done on how journalists change the practices of journalism by the influence of new media. For example, Lowrey and Mackay (2008) wrote an article about how journalists change their practices to address vulnerabilities in their occupation as they are challenged by bloggers (Lowrey and Burleson Mackay, 2008: 64). They argue that “editors’ awareness of local blogging activity corresponds to increased use of blogs as sources, discussion of blogs in planning meetings and adoption of the blogging form on news websites” (2008: 64). In this research, I’m interested in the way journalists look at new media practices and I’m especially interested in how they incorporate Twitter in their profession.

1

Someone who has a Twitter account is called a Tweep.

2

(6)

6

There is some research on how professional journalist can use Twitter as an active networking and news gathering tool, but not on how journalists tweet themselves. For example, Verbraeken writes about the benefits of journalists with a big Twitter network. On Twitter they can post questions and immediately approach their followers and ask for help (Verbraeken, 2009). Dutch foreign correspondent Olaf Koens points out that he gets 75% of his items from Twitter (Koens, 2008). He also thinks that Twitter in the future could be as indispensable as a good pub (Koens, 2008). But apart from what journalists say that they do with Twitter, there is a lack of empirical research on how (Dutch) journalists use Twitter. My research is an attempt to fill in part of this gap. I want to show how Dutch broadcast journalists use Twitter by analyzing their tweets.

The relevance of my research can also be shown through recent statements by the managing director of the NOS3, Jan de Jong. On 8 November 2010, he emailed all his employees regarding their use of social media. Shortly after, this internal memo was leaked on the Internet. In it, De Jong claims that social media, like Twitter, Facebook, LinkedIn and Hyves have potential risks. He points out that social media are public. When employees of the NOS post something on Facebook, Hyves or LinkedIn they have to keep in mind that it is very likely that there message becomes generally known; i.e. it becomes public information. The employee has to understand that what he or she writes will be seen as the opinion of the NOS in general even if the employee writes it in his or her own name. According to Jan de Jong, this can have a negative impact on the image of the NOS as a neutral, unbiased news organization. De Jong therefore asked all his employees to withhold themselves from announcements about or comments on internal affairs and on public broadcasting affairs. It does not matter if these affairs are already known by the audience. NOS employees cannot comment on matters of public debate that are related to public broadcasting and/or the NOS (Klöpping, 2010). Immediately after, a discussion among NOS employees (and other members of the public) evolved, because they could not tweet freely anymore. However, to what extent were Dutch broadcast journalists tweeting about internal affairs? Did the managing director have to set this straight or was it just a pre-emptive warning? By analyzing the tweets I gathered before this internal e-mail was sent, I may be able to address these questions.

To answer the question of how journalists tweet, I followed journalists of the public news broadcaster NOS and of the commercial broadcaster RTL. The aim of this study is to examine how these journalists are using Twitter (for their profession). The journalists selected for analysis are all those television journalists that actually appear on television. Therefore, their face and name are

3

(7)

7

linked by the audience to the news organization they work for. Are these journalists really withholding their comments on public debates that relate to their news organizations? Are they using it for story ideas, to get in contact with their audience or to promote their own stories? The central research question of this study is:

How do Dutch TV journalists of the NOS and RTL use Twitter? To answer this question several sub questions will be addressed:

 For what different purposes do the journalists use Twitter?

 What differences (and/or similarities) emerge between the tweets of journalists from RTL and the NOS?

(8)

8

Literature review

Journalism practice and the rise of the Internet

The rise of the Internet has had implications for the profession of journalists. The Internet offers new ways of collecting information. Journalists use online resources for background information to prepare interviews; to find sources or identify them; to check and verify facts; to read or watch news items of the competition; to become informed about current events and to get story ideas (Gunter, 2004: 90).

But the rise of the Internet has also meant new ways of reporting. Barrie Gunter (2004) argues that the use of the Internet by journalists will become more and more important in future news production. “Journalists will have to learn how to organize stories into structures conducive to interactive reading online. They might need to know not only about writing the basic news story, but also about effective use of audio, video, animation and databases that may form part of the larger interactive story package available to users” (Gunter, 2004: 90).

Internet can also be used as a communication tool. Several surveys that were conducted in the second half of the 1990s showed that most journalists used e-mail, but few saw it as a useful feedback tool. Journalists prefer telephone or face-to-face contact to e-mail conversation (Gunter, 2004: 99).

Changing position of the journalists

The arrival of the Internet has had more implications than new ways of collecting information and e-mail alone. Bardoel (2005) sees ‘network journalism’ as a fourth form of journalism, besides printed, radio and television journalism. It is interactive, multimedia oriented, hyper-textual and it is non-synchronically (Bardoel, 2005: 359 - 360).

Interactivity turned out to be one of the key elements of the Internet. It is possible to meet the needs and preferences of the individual user. Every user can get its own individual news that he or she wants to read. The power is no longer with the journalist, who supplies the information, but with the reader who searches for information (Bardoel, 2005: 360). David Carlson (2002) points out that traditional mass media lack two-way communication. Readers or viewers cannot interact with each other or with the journalists. On the Internet, this problem is easily solved because the medium itself is interactive (Carlson, 2002: 54).

(9)

9

In the more classical forms of journalism, the reader could only send a letter to the news organization. The way it is done on the Internet means a different relation between journalist and news consumer (Van Twisk, 2001: 37).

Van Twisk (2001) also points out that on the Internet there are no limitations in space and time. Everybody can access the news at any time of the day. For journalists deadlines belong therefore to the past, they can publish during the whole day. On the other hand, there is a permanent deadline (Van Twisk, 2001: 21). Journalists will still have the ambition to score a scoop. And because everybody can publish on the Internet there is a constant pressure (Pleijter, Hermans and Vergeer, 2007: 3).

Huub Evers (2002) sees problems with the rise of online journalism. He sees a tension between speed and reliability. The tension to publish immediately can be at cost of accuracy. Journalists are inclined to publish all information that is available, even when not all facts are checked. Because of the constant deadline on the Internet, there is minimal time for reflection or double-checking (Evers, 2002: 2). Van Twisk (2001) sees the same danger in the quick publication of news. He calls it “being premature”. The danger is that the pressure of the constant deadline will be at the cost of fair, accurate and unprejudiced reporting (Van Twisk, 2001: 27, 28). Thus, the reliability of news is placed in danger.

But journalist cannot turn their back on new technologies. Kevin Kawamoto wrote in 2003 that online news was becoming “a growing part of Americans’ news diet” (Kawamoto, 2003, x). With this in mind, Gunter quotes Singer who warned that “it is dangerous for journalists, whose job is deemed so crucial to society that is protected by our nation’s highest law, to stand on the sidelines as new media technologies emerge, take shape and spread” (Gunter, 2004: 105). Therefore, journalists should also familiarize themselves with technologies such as weblogs and Twitter, because social media are booming.

Journalism practice and the rise of weblogs

(10)

10

Blogging and citizen journalism

Through blogging the role of a journalist is changing, but also the rapidness of citizen journalism changes the practice of journalism. Stuart Allan (2006) wrote an article about the rapidness of citizen journalism and the blogging community after big disasters. He focuses on the inside information citizens can have in such situations. In his article, he uses the case studies of hurricane Katrina and the London bombings to explain: “Members of London’s blogging community were mobilizing to provide whatever news and information they possessed, in the form of typed statements, photographs *...+ and so forth” (Allan, 2006: 150). Bruns (2005) quotes Mitchell who argues that it is hard to beat a weblog when big news breaks. This became especially clear in the aftermath of September 11 and in the proceeding wars (Bruns, 2005: 175). “Because bloggers are closer to a story, they’ll often pick up the sort of things that traditional journalists miss. This is especially true for eyewitness blogs: blogs written by someone involved in a story” (Bruns, 2005: 175). Traditional media, like television have to consider the broadcast schedules and plan in an extra broadcast, where a blogger can post immediately pictures, videos or text online (Bruns, 2005: 175).

According to Rebecca Blood, a journalist and blogger, blogging is no new form of journalism.

“I’m not practicing journalism when I link to a news article reported by someone else and state what I think—I’ve been doing something similar around the water cooler for years. I’m engaged in research, not journalism, when I search the Web for supplementary information in order to make a point. Reporters might do identical research while writing, but research alone does not qualify an activity as journalism. Bloggers may point to reader comments as sources of information about the items they post, but these are equivalent to letters to the editor, not reporting. Publishing unsubstantiated (and sometimes anonymous) e-mails from readers is not journalism, even when it’s done by someone with journalistic credentials. Credible journalists make a point of speaking directly to witnesses and experts, an activity so rare among bloggers as to be, for all practical purposes, nonexistent” (Blood, 2003).

(11)

11

The blogging journalist

Blogging is also a thing journalists can adapt. Journalists can have good reasons to start blogging. They can post stories which would not get past their editors for several kinds of reasons (Bruns, 2005: 177). In this way, they escape from the limited space of the newspaper and can write about subjects that normally would not make the paper. This is particularly true for foreign correspondents who see their stories cut because of limited space (Hille and Baker, 2009, 10). Research among the websites of different English newspapers showed that foreign correspondents of Telegraph.co.uk were the first journalist of the paper that were offered a blog on the paper’s website. The foreign correspondents were ‘most frustrated about not having their articles published in the paper’ (Hermida and Thurman, 2008: 349). The same research showed that news executives made blogs available for journalists on the website because the executives wanted to retain control of journalists. Otherwise, the journalists may have created their own blog (Hermida and Thurman, 2008: 349).

Journalists can also expand the news they have broadcasted in the mainstream media on their blog (Grabowicz, 2003). In this way they use their blog as another medium for posting background material. And the blogs also allow journalists in this way to target a different audience, than the newspaper readers. A Mirror.co.uk editor explained that a certain journalist was happy with his blog because he “can address an audience in a way which he feels more comfortable with rather than having to dress everything up as a tabloid idea” (Hermida and Thurman, 2008: 349).

Overholser (2009) claims that journalists should not be looking at how they use weblogs as a ‘tool’ for journalism. The social media are here to stay and journalists have to look how journalistic values can be implied in the social media territory, rather than looking at how to use them for reporting (Overholser, 2009). Research by Hermida shows that at the BBC blogging has a greater impact on style, than on the substance of its journalism (Hermida, 2009: 3).

(12)

12

For journalists, this more personal style conflicts with the idea of a detached, objective observer (Heyboer, 2003/2004: 10). According to Hille and Bakker (2009), journalists could write more personal, but not all the journalists do this because they feel it is like crossing a line. Journalists rather apply the same journalistic standards to their blogs as they do to their other work. (Hille and Baker, 2009: 11). Applying a more personal style can have benefits for the reporters. They can invite the readers to actively participate on the weblog (Grabowicz, 2003). To get reconnected with their readers, journalists will have to be more transparent about how they do their job (Grabowicz, 2003). As more and more newspapers allow reporters and columnists to blog, the debate in newsrooms grew about how opinionated the blog is allowed to be (Heyboer, 2003/ 2004: 10).

An editor of telegraph.co.uk suggests that precisely this personal style and the emphasis on the personality of the writer, rather than the blog itself, may have contributed to the slow adoption by news sites. The emphasis on the journalist who writes a blog conflicts with the strongly established tradition in journalism that most reporting is written anonymously. The tone of a traditional journalist, who is mostly a fly on the wall that delivers information, is very different from the tone of most blogs (Thurman, 2008: 146).

Research by Hermida and Thurman of British newspapers showed that journalists do appreciate blogging and the “extra flexibility that the dialogue with readers have given them” (Hermida and Thurman, 2008: 349). Another survey showed that online journalists strongly hold on to the idea of interactivity, one of the key elements of blogging. Two-third of the Flemish respondents in that survey believed “that the future of online news production lies in interactivity” (Deuze et al., 2004: 22). In the Netherlands, 73% of the online journalists think that the best way to do online journalism is by building a stronger and interactive relationship with the audience (Deuze et al., 2004: 24).

But the whole idea of interactivity seems to be a farce. Interviews with US online editors and a content analysis of European online newspapers showed that reporters arehardly ever involved in debates with their audience (Domingo, 2008: 688). The same conclusion is drawn by Hermida (2009) who argues that journalists of the BBC like comments that readers can give on a weblog. The comment section becomes an area of debate, but the journalist rarely engage in the dialogue. They rarely intermingle and respond on their own weblog (Hermida, 2009: 12).

(13)

13

immediacy, publishing stories as quick as possible, and this fact strongly affected the development of interactivity” (Domingo, 2008: 692).

Transparent journalism

Bruns (2005) argues that there is a change noticeable in journalism, because of the rise of social media, like weblogs. For example, some reporters use participatory forms on the web to publish the complete text of their interviews. He calls this form of journalism ‘transparent journalism’ (Bruns, 2005: 176). Grabowicz also claims that weblogs “can give readers insight into the reporting process itself” (Grabowicz, 2003). Kovach and Rosenstiel even take a step further. Lasica (2003) quotes them when she argues that journalists must invite their audience in the reporting process. Many journalist who blog, expose the raw material of the story they are working on and ask readers for expert input. With this input, the journalist can write follow-up stories based on the readers’ tips and suggestions. For readers who blog and who get an opportunity to give feedback to journalists, this means an increase in loyalty and understanding (Lasica, 2003: 73,74).

According to Lasica, weblogs and journalism therefore depend on each other. Weblogs should be seen “as part of an emerging new media ecosystem – a network of ideas” (Lasica, 2003: 71). On the weblogs of readers journalists cannot find a complete stories or ideas. It is rather a community where bloggers discuss and reply on stories created by the mainstream media. A journalist must see it as a pool of tips, sources and ideas (Lasica, 2003: 71). Journalists think that this form of

“participatory journalism helps develop real community around reporters, stories, and the media company’s brand experience. With a weblog, for example, a reporter has a place to extend reporting, interact with readers, exercise personal conscience, and share some level of personality that might be absent from his ‘unbiased’ reports. These are elements that attract real community” (Bowman and Willis 2003: 55).

(14)

14

Journalism and Twitter

With the rise of Twitter a new micro-blogging tool was born. This has had implications for the journalism practice. “Traditional news operated on a 24-hour cycle. Blogs shortened this to minutes and hours. Twitter shortens it further to seconds. It's not right for every piece of information. It's certainly not well suited for longer analysis. But when it comes to instantly assembling raw data from several sources that then go into fully baked news stories, nothing beats it” (Ruffini, 2008). Ruffini also argues that Twitter changes blogging, but that there will always be a place for blogging because of the limited space (140 characters) journalists have on Twitter (Ruffini, 2008).

Twitter seems to be a popular tool for journalists who soon after its start saw the benefits of the social medium. “News organizations and reporters have been quick to adopt Twitter for an obvious reason: Its speed and brevity make it ideal for pushing out scoops and breaking news to Twitter-savvy readers” (Fahri, 2009). Hermida wrote something similar: “Twitter has been rapidly adopted in newsrooms as an essential mechanism to distribute breaking news quickly and concisely, or as tool to solicit story ideas, sources and facts” (Hermida, 2010: 3). Mainstream news reporters and bloggers are using Twitter to tweet the news as it happens and also add comments to it (Catone, 2008). Koens also mentions the rapidness of Twitter. “Moreover Twitter is really fast. An earthquake in China? I have read it more than 5 hours before it was covered in the Dutch media. A fire at the engineering faculty in Delft? I thought first it was a joke, because I could not find in the media. But it appeared to me that it just was not covered yet” (Koens, 2008). Verbraeken points out that Twitter is a quick newsfeed. “Twitter proved to be fast, after the plane crash at Schiphol airport. The Twitter community was very quick and reported live. Photos, comments and live streams could be found all over Twitter. The crash at Schiphol airport caused a news overload on Twitter *…+ Twitter is at its best when it is about ‘breaking news’. The riots in Mumbai in November of last year are a good example. Everybody knew what was happening thanks to the quick message flow on Twitter” (Verbraeken, 2009).

The use of Twitter by news organizations

(15)

15

headlines that rely on a short summary do not make any sense by themselves. This leaves users with no clue and it has little value for a news organization. A solution to this problem would be that journalists have to rewrite their headlines, so that they make sense on Twitter. The only downside to this is that it takes time (Thornton, 2009).

But there are also news organizations that try new things on Twitter. Tenore sees potential of Twitter in the way it can help journalist reporting while they are in the field. “That could be during an event -- let's say the music critic uses Twitter at a big concert or festival to send tweets and receive them from readers at the show ... Same, say, with a sportswriter or one of our columnists, either at an event or just day to day.” (Tenore, 2007a).

A similar use of Twitter is described by Pam Mapels in an article of Bruno Giussani (2008). Giussani asked the managing editor of the St. Louis Post-Dispatch, Pam Mapels, how they incorporated Twitter in a news story. The newspaper dedicated a special part of the site to the topic of the two-year long construction shutdown of a highway near St. Louis. That topic was highly relevant to the community because the highway was used a lot and the situation would differ from day to day. The Twitter project was a success. Alongside journalists of the newspaper worked 11 citizen reporters. They incorporated the tweets on the website (Giussani, 2008). In this example, Twitter was also used by (citizen) reporters in the field. According to Maples, not every story is suitable for Twitter coverage: “Twitter has specific strengths that make it an asset in covering certain types of stories -- where immediate micro-information is very relevant to the readers -- while in other cases, it might not be worth taking the time to add it to your coverage arsenal” (Giussani, 2008).

Another use of Twitter by news organizations is to send out breaking news. In 2007, a US local newspaper, the Nashua Telegraph, adapted the Twitter format for breaking news. The tweets with breaking news included a tinyurl address linking followers back to the Nashua Telegraph’s website. The managing editor of the website of the newspaper, Damon Kiesow, explained that some papers use Twitter for their whole news feed, but according to him that is an inappropriate use of Twitter. The breaking news feed is selective, so users will not get an overload of tweets. They select only the breaking news “that’s important regardless of where you are” (Oliver, 2007).

(16)

16

Use of Twitter by journalists

Koens, a Dutch journalist, wrote in September 2007 the following about Twitter: “What a useless medium” (Koens, 2008). In January of the next year, his opinion changed: “When I look in my journalistic agenda I see that 75% of my upcoming writings have to do with Twitter in one or another way” (Koens, 2008). Van Bijnen (2010) wrote something similar: “You can get in touch with useful sources on Twitter. I have found several via the message service. Eventually you have to call them or visit them. Twitter makes the life of a journalist a lot easier” (Van Bijnen, 2010). Thornton also writes about the usefulness of Twitter for journalists, as he states: “Journalists can find success on Twitter by crowd sourcing story ideas and stories, connecting with sources, doing research and more” (Thornton, 2009). As these examples show, Twitter seems to be more interactive when individual journalists use Twitter. They do not only use the social medium to push out scoops, but also to find sources.

Kirkpatrick, writer for the popular technology weblog ReadWriteWeb.com, describes how ReadWriteWeb uses Twitter for their website. They discover big items of interest first on Twitter and also tech news tips come to them via Twitter. They use it to arrange interviews and ask interview questions via Twitter, but also they receive question suggestions of followers. Another way they use Twitter is to perform public interviews. By putting out questions to their followers, the journalists of ReadWriteWeb get more research than they could get by calling people by phone. They realize that the people who answer their questions are not generally representative of the whole population, but for qualitative interviews, Twitter is a very useful tool. Twitter can also work very well for finding quick answers to small tech questions and for tech support (Kirkpatrick, 2008). Patrick Ruffini (2008) also admits that Twitter is “the first place I turn to if I need a quick question answered, and for raw political intelligence” (Ruffini, 2008). Another thing ReadWriteWeb uses Twitter for is quality assurance. They get feedback from their followers on misspellings, missed links and other publishing mistakes, via Twitter (Kirkpatrick, 2008). The last way ReadWriteWeb uses Twitter is to promote their online articles. They claim that this “is probably the crassest way a journalist can use the medium” (Kirkpatrick, 2008). Twitter is a way to get more followers to read their posts on their weblog and get more traffic to their website.

(17)

17

the last point, journalists can use Twitter to share their opinions and to give the public a sense of their personality (Smith, 2009). If you compare the interactivity between blogs and Twitter, it seems that on Twitter journalists have much more interaction with their readers. Blogs on the other hand tend to be used as a means of publishing stories that would have otherwise not been published. Thus, Twitter seems to fill a different function than blogging.

Twitter and interaction

Similar to weblogs, the buzzword of Twitter is interactivity, but on Twitter maybe this is not just a buzzword. This study will examine if journalists actually are interacting with their audience via Twitter. “Conventional wisdom says that to be good at using social media sites like Twitter, one must be social” (Thornton, 2009). The interaction on Twitter is something Julie Posetti claims to be a good thing for the relationship between journalist and followers/news consumers, as she states: “Increasing numbers of professional journalists are signing up to Twitter, viewing the site not only as a reporting tool but also as a device for live audience interaction. One crucial effect of this interaction has been to break down those barriers between news producers and consumers that have tended to isolate the more traditional ‘legacy media’” (Posetti, 2009). Mary Jean Tenore also argues that Twitter can connect users to a news website who would otherwise not read a newspaper or visit that papers website (Tenore, 2007b). The profit journalists gain from this new form of interaction, are new connections. “As journalists, we thrive off of connections – connections to the communities we cover, to the stories we write and to others in the field who can help us develop our careers” (Tenore, 2007b). Paul Bradshaw also argues that contacts are central to the work of a journalist. He says Twitter makes it easier to find them. “It’s where blogging meets social networking” (Bradshaw, 2008). But to get many connections and be social, a journalist must be a bit personal and not just a neutral observer. Thornton gives the example of David Pogue, a New York Times columnist, who has about 850,000 followers. Thornton thinks this is partly because “he is entertaining and personal, while also interacting with fans” (Thornton, 2009). He also claims in the same article that it requires a lot of time to be interactive on social media. But on the other hand, this is an investment a journalist must make in order to establish good connections and make the most out of Twitter (Thornton, 2009). For example, a German journalists for the newspaper Handelsblatt explains how his connections make his work easier: “One day, when I was researching an article, I was looking for a specific piece of information *…+ So I just asked the question on Twitter – and had the answer I was looking for within minutes” (Luchsinger, 2008).

(18)

18

communication. Twitter seems to fill this gap, but are Dutch journalists also interacting with their audience on Twitter? Do they have a big network? Are they showing their personality or just posting the latest news? And how are they interacting with other journalists? Twitter cannot only establish the connections between citizens and journalists but also the connections between different journalists. According to Tenore, journalists are always looking for new ways to communicate with each other (Tenore, 2007b). It even can help them find a job. When they answer Twitter’s question “What are you doing?” with “Looking for a job”, the British journalism site, journalism.co.uk sends them job updates (if they follow the special Twitter account the site has) (Tenore 2007a).

Ethical standards

Through the use of Twitter, some ethical norms are changing. Fahri mentions this when he quotes Mark Briggs: “Whatever you put out there doesn’t have to be triple checked, but it can’t be reckless or inaccurate, either” (Fahri, 2009). Harris, a software engineer at the New York Times, sees a risk, as she states: “I suppose there might be a risk if journalists leaked internal information accidentally via Twitter, but these risks are not unique to the medium and I think it might be cool in terms of transparency if more journalists did Twitter both on the stories they're covering or their general lives and interests" (Tenore, 2007a).

Hermida (2010) also argues that “there are indications that journalism norms are bending as professional practices adapt to social media tools such as micro-blogging. *…+ Journalists apply normative news values to determine if a specific tweet is newsworthy.” (Hermida, 2010: 4). In other words, the gatekeeper role is enforced because journalists filter their tweets. They publish the tweets they think are newsworthy. But these tweets are according to Posetti blurred tweets that contain public and private information. Twitter “merges the professional and the personal, the public and the private — blurring the lines of engagement for journalists trained to be didactic observers and commentators rather than participants in debates and characters within stories. Reporters’ use of the platform to express feelings and opinions on a range of issues has raised red flags about professional conduct and bias” (Posetti, 2009). The question is how opinionated the tweets of the Dutch broadcast journalists are. Are they mixing fact and opinion? Do they show their personality or are they using Twitter just another news outlet?

(19)

19

expression and not a journalistic practice. The board explained that whether a tweet is a journalistic practice depends on the content of the tweet (Villamedia, 2010). Even if a journalist is tweeting, that does not make it a journalistic practice as such. Apparently, there is a distinction between the different types of tweets.

Depth of the reporting

(20)

20

Research Design and Methodology

To answer the question how Dutch journalists of the NOS and RTL use Twitter, I will adopt a qualitative research design (Creswell, 1998; Strauss and Corbin, 1990). Qualitative research is a type of research “that produces findings not arrived at by statistical procedures or other means of quantification” (Strauss and Corbin, 1990: 11). The bulk of the analysis is interpretive. The purpose of this analysis is to discover concepts and relationships in raw data (Strauss and Corbin, 1990: 11). According to Strauss and Corbin qualitative methods can be used to explore areas about which little is known (Straus and Corbin, 1990:11). As described in my literature review not much is known of the actual use of Twitter by journalists. Therefore, a qualitative research design is justified. As a means of investigating journalists’ tweets, a content analysis that employs both inductive and deductive coding categories was employed as the primary method for examination.

Population, sample and sampling procedures

In order to analyze how journalists use Twitter, I have chosen to follow 73 Dutch broadcast journalists from RTL and the NOS. The journalists consisted of 32 journalists from the NOS and 38 from RTL. I followed all the reporters, correspondents and anchors of both news organizations that had a Twitter account in September 2010. I did not follow the editors because of two reasons. The first reason is a practical one; I would gather too much data. The second reason is that journalists that work in a newsroom do not have a familiar face. They are unknown by most of the viewers of the news broadcasts. The correspondents, reporters and anchors on the other hand are very visible and known for their journalistic work by the audience and therefore are likely to have more followers. The following journalists are included:

NOS Reporters

1. Ron Fresen Political reporter

2. Dominique van der Heyden Political reporter 3. Jeroen van Dommelen Political reporter 4. Xander van der Wulp Political reporter 5. Hans Andringa Political reporter

6. Lex Runderkamp Reporter

7. Jeroen Wollaars Reporter

8. Marieke de Vries Reporter

9. Martijn Bink Reporter

(21)

21

11. Pauline Broekema Reporter

12. Gert-Jan Dennekamp Reporter

13. Gerri Eickhof Reporter

14. Theo Verbruggen Reporter

15. Marjolein Hogervorst Reporter

Correspondents

16. Wessel de Jong Correspondent in Brussels 17. Chris Ostendorf EU correspondent in Brussels 18. Arjen van der Horst Correspondent in London 19. Saskia Dekkers Correspondent in Paris 20. Peter ter Velde War Correspondent

21. Kysia Hekster Correspondent in Moscow

22. Ron Linker Correspondent in Washington, DC 23. Eelco Bosch van Rosenthal Correspondent in Washington, DC 24. Rop Zoutberg Correspondent in Madrid

25. Andrea Vreede Correspondent in Rome 26. Hennah Draaibaar Correspondent in Paramaribo 27. Sander van Hoorn Correspondent in Tel Aviv 28. Robert Portier Correspondent in Sydney

29. Michel Maas Correspondent in Jakarta

30. Kees Broere Correspondent in Nairobi

Anchors

31. Sacha de Boer

32. Rik van de Westelaken 33. Astrid Kersseboom 34. Jeroen Tjepkema 35. Herman van der Zandt RTL

Reporters

1. Jos Heymans Political Reporter

2. Frits Wester Political Reporter

3. Lotte Ragut Political Reporter

4. Fons Lambie Political Reporter 5. Helma Huizing Political Reporter

6. Geert Gordijn Reporter

7. Hans Schutte Reporter

8. Hella Hueck Reporter

9. Rik Konijnenbelt Reporter

10. Jaap van Deurzen Reporter

11. Koen de Regt Reporter

12. Martin van Norel Reporter

13. Floor Bremer Reporter

14. Suzanne Aalers Reporter

15. Merijn Doggen Reporter

16. Pim Sedee Reporter

17. Eva Kropman Reporter

(22)

22

19. Sandra Schuurhof Reporter

20. Jeroen Wetzels Reporter

21. Steven Schoppert Reporter

22. Betty Glas Reporter

23. Iris van Delden Reporter

24. Roel Geeraedts (Investigative) reporter 25. Hester van Yperen (Investigative) reporter

Correspondents

26. Jeroen Akkermans Correspondent in Berlin 27. Pauline Valkenet Correspondent in Rome 28. Arthur de Leeuw Correspondent in Nairobi 29. Erik Mouthaan Correspondent in New York

30. Tim Dekkers Correspondent in Sydney

31. Nina Jurna Correspondent in Paramaribo 32. Marije Vlaskamp Correspondent in Beijing 33. Reinoud Broekhuijsen Correspondent in Russia

Anchors 34. Antoin Peeters 35. Suzanne Bosman 36. Femke Wolthuis 37. Jan de Hoop

38.

Daphne Lammers

In the previous exploratory case study, I focused on two disasters that happened in the year 2010 and caused a lot of tweets. The events happened in the two countries where the correspondents that I followed lived. In that way I could compare the different Twitter activities of the correspondents about the same topic. It turned out that I only used the disaster to set a timeframe. I did not link any of the conclusions to the aftermath of the actual disasters. Therefore the selection of data in this study was based on a timeframe. The journalists were followed for a two week period, from the 13th to the 27th of September. All tweets posted by the above journalists during this time were included in the analysis. Initially, I wanted to use a one month period, however, after two weeks, I already gathered 4069 Tweets. Therefore, the timeframe was limited to two weeks due to time constraints; one month of capturing would have produced too much data to manage.

(23)

23

Instrument / method used and data analysis

For my research I used a qualitative content analysis with a deductive and inductive part. The deductive part of the analysis elaborates on the theory that I have conducted during a previous research. During this study, I analyzed the tweets of four correspondents in the aftermath of a disaster. I found six different categories of tweets: news, background, journalistic work, promotion, interaction and personal tweets. The most common category in which the journalists tweeted was news tweets; however, the second most differed for all the journalists. Therefore, I wanted to broaden this study and see if there are more similarities in the second category if my sample size is bigger. I’m going to use these six different categories in my new research as deductive categories. In my previous research these categories cover most of the tweets. But I have to refine the categories. For instance the category of interaction has to be refined. In my previous research, tweets of journalist to average followers and tweets to other journalists all were collected in the same category. The personal tweets are another category that has to be revised. In my previous research, tweets about the personal lives of journalists and tweets with personal comments on the news belonged to this category. In this study, I will define the categories better and accomplish them with the categories I have found in the inductive part of my study.

For the inductive part grounded theory will be used to analyze how 73 journalists (reporters, correspondents and anchors) of RTL and the NOS use Twitter. A grounded theory is derived from data that is systematically gathered and analyzed through the research process. With a grounded theory approach the researcher begins with an area of study (the data) and allows the theory to emerge from the data (Strauss and Corbin, 1990: 12). “The intent of a grounded theory study is to generate or discover a theory, an abstract analytical schema of a phenomenon, that relates to a particular situation” (Creswell, 1998: 56). During the analysis several rounds of coding will be conducted, during which coding categories will be created, modified and discarded via the use of feedback loops.

(24)

24

In the literature, I have found on the grounded theory approach the most mentioned data that can be analyzed come from interviews and observations (Strauss and Corbin, 1990: 20). So my grounded theory study differs a bit because I’m not doing any interviews, my focus is on tweets instead. But my previous exploratory case study proved that the tweets are textual data that is suitable for a grounded theory study. Moreover, similar studies on UGC (namely discussion forums) have adopted a grounded theory approach (see e.g. Van Zoonen, 2007).

There are advantages and disadvantages by analyzing tweets instead of doing interviews and analyzing the transcripts. An advantage is that I can directly analyze how the journalists use Twitter instead of analyzing how the journalists say they use Twitter. There can be a difference in their actual tweeting habits and the way they say they use Twitter. In an interview situation, the journalist may give socially acceptable answers or romanticize the use of Twitter. A disadvantage is that I cannot ask the journalists why they tweet what they tweet. I can analyze how they use Twitter, but not why they choose use Twitter in this way. Because my research question focuses on how journalists use Twitter, I chose to focus on the tweets only.

To analyze the tweets of the 73 Dutch journalists, I used the first coding phase as described by John Creswell (1998) to discover a pattern. Some tweets will fall into a category that I already developed in my previous research. Still it was useful for me to label all the tweets because I wanted to redefine the categories. For example the tweets that will fit into the ‘news’ category were easily to discover, but I added for instance if the news was from the hand of the journalist or that he copied it from other media. Other tweets that did not fit into any of the previous categories had to be open coded, labeled and categorized from the beginning. I used an inductive approach to those tweets.

In the open coding phase, the researcher takes the data and segments it into categories of information. According to Creswell, you have to develop a number of categories, slowly reducing the number to 5 or 6 categories that become major themes in the study (Creswell, 1998: 241, 242). Or as Strauss and Corbin describe it “open coding is the part of analysis that pertains specifically to the naming and categorizing of phenomena through close examination of data (Strauss and Corbin, 1990: 62). In this phase, I have to give every tweet a specific label and after that categorize them in a number of categories. When I have categorized both inductive and deductive categories I use feedback loops to modify the coding categories of both approaches.

(25)

25

from the labels 11 different categories were specified in which the journalists tweet. Three of those categories are divided in subcategories which go deeper in to the results.

Archiving, organizing and managing the data

On the website of Twitter you cannot go back in time endlessly (anymore). Therefore, I chose to follow the journalists for two weeks that were in the future when I started with my thesis. Every other day, I visited all the Twitter accounts of the 73 journalists and saved their new tweets in a Word document. When I had gathered all the tweets I printed them hardcopy. All tweets are available in appendices II and III of this thesis.

I coded the tweets by hand and managed to get three times trough all the data. First I read all tweets and also looked if I could trace back the tweets on which the journalists replied. If I could I also added these tweets to the list. I also tried to define who the person was to whom the journalist replied. I also added this information to my Word document. After doing this I read the tweets again and kept a new document in which I wrote down which label belonged to which tweet. The last time I went over these Tweets I again looked at the labels and the tweets and if I still thought this was the correct label I registered it in an excel document. The tweets of which I had developed another point of view, I could adjust during this third time I went over all the tweets and fine tune the labels.

Reliability and validity

This research and the first ideas are checked by my supervisor and fellow students. To increase the validity of my research I asked assistance from my supervisor. According to Creswell and Miller, a ‘peer debriefer’ provides “support, plays devil’s advocate, challenges the researchers’ assumptions, pushes the researcher to the next step methodologically, and asks hard questions about methods and interpretations. *…+ This procedure is best used over time during the process of an entire study” (Creswell and Miller, 2000: 129).

(26)

26

discuss my thesis subject with my classmates and made a research design which was approved by the teacher of the thesis class and my supervisor.

Intra-rater reliability

Scott’s Pi Number of categories

1.0 2

Greater than 0.9 2

Table 1: Intra coder reliability measured with Scott’s Pi

To check if I did my analysis right I use an intra-rater reliability test to check myself for stability. After labeling the tweets I recoded 10% of the total amount of tweets again for intra-rater reliability. I selected the tweets that I labeled again at random. I relabeled 413 tweets. I picked at random 255 tweets of RTL and 158 tweets of NOS, since the journalists of RTL had produced a lot more tweets in the period that I followed them. The intra-rater reliability was calculated in two ways: by measuring the percentage of agreement and Scott's Pi. The last is a statistical method that measures what would be the chance that I coded the label per accident in the right category. Table 1 shows the intra coder reliability for the different categories. All the codes scored greater than 90% on the percentage of agreement calculation and 10 categories scored greater than .7 with regard to Scott's Pi. Neuendorf (2002) argues that this is an acceptable level of reliability, because reliability above 60% agreement with Scott’s Pi greater than .3 is an acceptable level of reliability (Neuendorf, 2002). Although Neuendorf refers to inter-rater reliability, the same standards seemed reasonably acceptable for my intra-rater reliability test. One category was not checked in this reliability test. No tweets of the category ‘other’ were relabeled. It was a very small category and no tweets of this category showed up in my sample, which was taken at random. Because no conclusions were drawn upon this category I did not consider this as a problem.

Ethical considerations

(27)

27

read by everyone. When it’s about researching the tweets Jankowski and Van Selm (2007) quote Bruckman who suggests that Internet-based material may be freely cited when:

 It is officially, publicly, permanently archived;

 No password is required to archive access;

 No site policy prohibits it;

 The topic is not highly sensitive (Jankowski and Van Selm, 2007: 282).

All these points are applicable to the tweets are used for this study. The tweets are officially, publicly and archived for a period of time. You can go back in time and see older tweets without registration or passwords. Twitter.com doesn’t prohibit that you save tweets and use it for research. The topics I investigate are not highly sensitive, at least not for the journalists of whom I investigate the tweets. The tweets can be personal, but they are sent out to all their followers and the journalists have many followers. On September 22, 2010 the 70 journalists had 175,836 followers all together4. They are only the followers who chose to get updates from the journalists. People do not have to follow the journalists to read their tweets. There is one exception. Eva Kropman working for RTL has protected her tweets. You can only read those if she allows you to follow her. But still you do not need any password to read her tweets. You only have to have a Twitter account and permission from her to follow her. All the other journalists do not protect their tweets. Everyone who would like to can read them, even if they do not have a Twitter account themselves. Therefore, I can use the tweets without needing any permission.

The tweets of the journalists of the NOS are not directly linked to their news organizations. Some journalists do refer in their name that they are working for the NOS like: @MariekeNOS, @ebvrNOS and @nosnoord. Of the 35 journalists that I followed working for NOS, only 6 of them referred to the news broadcaster in their name. At RTL 12 journalists of 38 put RTL in their Twitter names, for instance: @MerijnRTL, @Geert_RTLNieuws and @RTLnieuwsbetty. By doing this they show that they are working for RTL or NOS. Most of the RTL journalists also have the logo off the news program placed in the avatar of their Twitter account. On the homepage of RTL news is a Twitter feed embedded with the latest Tweets of the reporters and editors placed (RTL.nl, 2010) The journalists off RTL are thus indirectly publishing on their news website and maybe therefore less free in what they tweet. Also the journalists of NOS aren’t completely free in what they can tweet. The NOS has a policy for its journalists and the use of Twitter (Stekelenburg and Overdiek, 2009). So the journalists are highly aware that everybody can read what they tweet.

4

(28)

28

Limitations

My research has some limitations. First, a possible limitation could be the timeframe used. Because the two weeks I have chosen to follow the journalists was set in the future it could happen that nothing exciting occurs. If that would be the case the journalists would not have much to tweet about. But I was pretty sure this was not going to happen, because on the 9th of June 2010 there were elections and by the time I started with my thesis in September still no government had been formed. So at least there would be some political news.

The second limitation also has to do with the timeframe. In the two weeks I have gathered tweets, there is a big difference in the amount of tweets that each journalist produces. There are journalists that tweet very little or have an account but do not post anything (anymore). This is not a limitation per se, because it also tells me something about how the different journalists use Twitter. But because I only follow them for two weeks it could be that a journalist is on a holiday or has an illness that I do not know about. Then his or her use of Twitter is affected, but I should be able to see this when I visit the Twitter account later in time and see that the journalist is then very active. Some journalists also announce that they are going to be ‘offline’ for a while because they are on a holiday.

Further, I can only see if the journalist got reactions at a certain question if the reporter commented back on a specific answer. I cannot see what the reporter sees on his Twitter account. All the messages he gets but on which he does not react, were not visible for me. The only way to cover this problem was when I would follow every follower of the journalists and filter all the times they mention the name of the journalist. But this was not doable. Altogether the journalists have hundreds of thousands of followers, which I could have never followed all.

Another limitation can be that I have a lot of knowledge on the subject because of my previous exploratory. This can cloud my thoughts when I am labeling and categorizing the tweets, but on the other hand is my previous research is a good thing. I know what I did wrong the previous time. I can refine the categories I have found before and build further on the knowledge I already have.

(29)

29

Results

During the two weeks of following the reporters, correspondents and anchors I gathered 4,069 tweets. The 35 NOS journalists twittered in total 1,541 times and the 38 RTL journalists shared 2,528 tweets. The first striking difference is that the RTL journalists produced almost a 1000 more tweets than their colleagues at the NOS. Not all journalists twittered the same amount of tweets. Some did not tweet at all.

NOS

Reporters Amount of tweets

1. Ron Fresen 48

2. Dominique van der Heyden 124 3. Jeroen van Dommelen 17

4. Xander van der Wulp 74

5. Hans Andringa - 6. Lex Runderkamp 169 7. Jeroen Wollaars 126 8. Marieke de Vries 204 9. Martijn Bink - 10. Rienk Kamer 16 11. Pauline Broekema - 12. Gert-Jan Dennekamp 1 13. Gerri Eickhof - 14. Theo Verbruggen 35 15. Marjolein Hogervorst 64 Correspondents 16. Wessel de Jong 20 17. Chris Ostendorf 5

18. Arjen van der Horst 151

19. Saskia Dekkers 25

20. Peter ter Velde 42

21. Kysia Hekster 69

22. Ron Linker 31

23. Eelco Bosch van Rosenthal 147

24. Rop Zoutberg 42

25. Andrea Vreede -

26. Hennah Draaibaar -

27. Sander van Hoorn -

28. Robert Portier 74

29. Michel Maas 18

(30)

30 Anchors

31. Sacha de Boer 3

32. Rik van de Westelaken 33

33. Astrid Kersseboom -

34. Jeroen Tjepkema 3

35. Herman van der Zandt -

At the NOS, there were 25 journalists who actually twittered during the two weeks of investigation. But among those 25 there were also journalist who twittered infrequently. Four journalists posted less than a tweet per day. Of the 35 journalists that I followed, 21 actively twittered during the period. Also at RTL, not all journalists used Twitter actively.

RTL Reporters 1. Jos Heymans 46 2. Frits Wester 424 3. Lotte Ragut 11 4. Fons Lambie 98 5. Helma Huizing 12 6. Geert Gordijn 13 7. Hans Schutte 45 8. Hella Hueck 127 9. Rik Konijnenbelt 58

10. Jaap van Deurzen -

11. Koen de Regt 203

12. Martin van Norel 4

13. Floor Bremer 8 14. Suzanne Aalers 1 15. Merijn Doggen 4 16. Pim Sedee 27 17. Eva Kropman 54 18. Britta Sanders - 19. Sandra Schuurhof - 20. Jeroen Wetzels 7 21. Steven Schoppert 27 22. Betty Glas -

23. Iris van Delden -

24. Roel Geeraedts 454

25. Hester van Yperen 2

Correspondents

(31)

31 27. Pauline Valkenet 24 28. Arthur de Leeuw 44 29. Erik Mouthaan 106 30. Tim Dekkers - 31. Nina Jurna 1 32. Marije Vlaskamp - 33. Reinoud Broekhuijsen 19 Anchors 34. Antoin Peeters 15 35. Suzanne Bosman 1 36. Femke Wolthuis 118 37. Jan de Hoop 440 38. Daphne Lammers 29

At RTL, 31 journalists posted tweets during the 15 days that I followed them. However, 11 journalists posted less than 15 tweets, averaging less than a tweet per day. That said, 20 journalists used Twitter actively, which is comparable to the 21 journalists at the NOS who twittered on average at least once a day. What is striking is that the RTL journalists, who used Twitter actively, produced more a lot more tweets than their colleagues at the NOS. Three RTL journalists even posted more than 400 tweets in two weeks, while the most active journalist at the NOS posted ‘only’ 204 tweets.

Interaction

I

nteraction was divided into different subcategories in order to provide more insight into how journalists interacted. The different types of interaction can be seen in Graph 3.

Graph 3: Different kinds of interactions of all journalists together

The first group of people journalists interacted with was colleagues. There are three different types of colleagues: colleagues from the same news organization, colleagues that worked for other media and colleagues that work at the technical side.5 Also the kind of interaction the journalists had with their colleagues was registered. In Appendix I, you can find the extended list of 208 labels that were created, including the different kinds of interaction. For example, a journalist interacted with a colleague of the same news organization about current affairs, or was answering a question.

5

Because these journalists work for broadcast media, they have interaction with for instance cameramen or television directors. Because these are mainly freelancers who work for different bidders, I have not subdivided the technical staff between the different news organizations and other media.

4% 10% 22% 3% 3% 2% 2% 38% 1% 13% 2% 0%

Interaction at both news organizations

Interaction to thank somebody Interaction colleague NOS

Interaction colleague (other than NOS) Interaction coworker technique Interaction politician

(35)

35

Another group of people journalists interacted with were politicians, spokesmen and political activists at the national, regional or local level. Regarding the latter, people that were active in local politics or persons that were active in political groups were coded as ‘political activists’. Most of the time they had a job beside their political activities, but when they interacted with journalists it was mostly in the light of their political interests. Interaction with spokesmen can be any kind of spokesmen. Mostly the spokespersons were political informants, but they were also lobbyists included. This distinction was not made clear in the labels because all spokesmen have the same goal. That is to represent their company or party the best way as possible. They have always a special interest in their interaction. For interactions with these groups also was included what the interaction was about. The subjects of these interactions can be found in the extended list of labels in appendix I. It ranges from joking with a politician to interaction with a politically active person about the journalistic work or asking a spokesperson a question.

Another kind of interaction was interaction with random followers. These followers were people that could not be related to the journalist in any way and did not have any of the occupations as described above. Some of the journalists had many followers because they had a familiar face with the public. These followers did not know the journalist personally, but did follow him, for instance to get the latest news or to ask questions. Interactions with random followers can be found in appendix I and consist of for example: answering a question that does an appeal to the journalists expertise, interaction with a random follower about the personal life of the journalist and interaction with a random follower about their personal taste in for example television programs, books and so on.

Journalists also interacted with acquaintances. These were interactions with people the journalist seemed to know. This presumption was based on the fact that the journalist followed the person back who he was interacting with. Therefore the journalist must know the person, or has had interaction before. Most of the time it was not clear what the relation was between the journalist and the acquaintance, other than that they must know each other. The sort of interaction the journalists had with acquaintances was not relevant because the interaction with this kind of follower. It quickly became small talk.

(36)

36

could visit them. The interactions with several people at once were hard to label because a journalist sometimes replied to both a journalist and a random follower who had the same question. Therefore, the followers with whom the journalist were interacting were not registered in these labels. Otherwise one tweet would have more than one label and this would blur the results of this research. After all, it was not very important because the amount of tweets that got this label is marginal.

The last sort of interaction was the interaction where the journalist thanked somebody. The journalists thanked many people for several kinds of reasons, but often it was hard to trace back why journalists thanked that specific person. Moreover this sort of interaction was most of the time more some of politeness than a real interaction. Therefore the person to who the gratitude was shown was not registered.

Interaction at RTL

(37)

37

Graph 4: Different kinds of interaction at RTL

The second biggest category of interaction at RTL was interaction with colleagues that did not work for RTL. Again Jan de Hoop scores high (32% of his interaction was of this kind). He tweets a lot with colleagues of a Dutch radio station, Q-music. He gets up very early in the morning and listens to a show on that music station. He uses Twitter to say “Good morning” to the colleagues who also have to get up very early.

The third category of interaction is interaction with an acquaintance. Roel Geeraedts, an investigative reporter, twittered a lot with people he knew. Of his tweets that contained interaction, 25% were directed at somebody he already follows. Femke Wolthuis, another anchor of RTL, also twittered a lot to acquaintances; 32% of her replies were to somebody she knew. Both journalists interacted more with people they knew, while interacting less with random followers.

4% 9% 20% 3% 3% 0% 2% 42% 1% 14% 2% 0%

Interaction RTL

Interaction to thank somebody Interaction colleague RTL

Interaction colleague (other than RTL) Interaction coworker technique Interaction politician

Interaction policitally active person Interaction spokesman

(38)

38

Interaction at NOS

Graph 5 below shows that the biggest category of interaction at NOS was also interaction with random followers. When journalists replied 30% (204 tweets) was meant for random followers. Again some journalists had more contact with random followers than others. Of the replies of Dominique van der Heyde, a political reporter, 48% was meant for random followers. Lex Runderkamp posted 103 interactive tweets and 30% of those were replies to people he did not know in person. And Eelco Bosch van Rosenthal, correspondent in America, replied in 48% of his interactive tweets to random followers.

Graph 5: Different kinds of interaction at NOS

The second biggest subcategory at the NOS was the interaction with colleagues other than NOS, thus colleagues that work for other media. The difference between this category and the biggest one is only 2%. Journalists of NOS that interact a lot via Twitter seem to give random followers an equal amount of attention as colleagues. For example Jeroen Wollaars, a reporter, interacted in 101 of his

3% 12% 28% 3% 4% 4% 2% 30% 0% 11% 2% 1%

Interaction NOS

Interaction to thank somebody Interaction colleague NOS

Interaction colleague (other than NOS) Interaction coworker technique Interaction politician

Interaction policitally active person Interaction spokesman

BREAKING: #J OURNALISM @T WITTER

BREAKING:

#J

OURNALISM

@T

WITTER

A

STUDY ON HOW JOURNALISTS AT

RTL

AND

NOS

USE TWITTER

Esther de Kloe

Student number: 1912089

Master Thesis Journalism Studies

Rijksuniversiteit Groningen

August 2011

Acknowledgements

Table of contents

Introduction

Literature review

Journalism practice and the rise of the Internet

Changing position of the journalists

Journalism practice and the rise of weblogs

Blogging and citizen journalism

The blogging journalist

Transparent journalism

Journalism and Twitter

The use of Twitter by news organizations

Use of Twitter by journalists

Twitter and interaction

Ethical standards

Depth of the reporting

Research Design and Methodology

Population, sample and sampling procedures

38.

Instrument / method used and data analysis

Archiving, organizing and managing the data

Reliability and validity

Intra-rater reliability

Ethical considerations

Limitations

Results

Categories

Interaction

I

Interaction at both news organizations

Interaction at RTL

Interaction RTL

Interaction at NOS

Interaction NOS