Latest News Updates on the Web 1

Running head: LATEST NEWS UPDATES ON THE WEB

Latest News on the Web: Content Change and News Topic and Type Jin Xu Department of Communication Studies Winona State University August 25, 2008

Jin Xu PAC215 Winona State University Winona, MN 55987 507-457-2267 [email protected]

Latest News Updates on the Web 2

Abstract The research examines content change in CNN.com’s latest news and how it is different regarding news topics and news types. Real-time updates to 228 randomly selected stories from June to August, 2007 were content analyzed. It has produced first-hand evidence about real-time online news updates with regard to news topics and news types. The findings suggest that timely content is a distinct hallmark of the latest news at CNN.com, that the updates may provide “substantial background information,” and that users’ interest in disruptive and episode-oriented news coverage are catered to by content change. It is concluded that the question of when to get the news online makes a difference.

Latest News Updates on the Web 3

Latest News on the Web: Content Change and News Topic and Type With promises of information anytime, anywhere, and in any form,1 the Web is leading toward a period of transformation. As far as Web-based news is concerned, one of the transcendent advantages is ease of content change, making it a preferred real-time medium for breaking news. It is also possible to repackage and constantly update news stories.2 This affects news production and news consumption, raising expectations about the speed of coverage.3 The emphasis on timeliness underscores the most fundamental change that online capability brings to the news business.4 Breaking news in combination with deep insight can feed the audience with round-the-clock content, an Internet analyst at Nielsen/NetRatings said.5 The Web has been changing news consumption and production. Accepted as an alternative source, online news has gained popularity.6 Breaking news and news updates are routinely read.7 News production has been undergoing change as well. It is believed that as online news operations mature in managing a real-time, 24-hour news environment, online journalism will provide increasingly compelling news content.8 The changes have already been reflected in shorter news cycle. Some sites put “update tickers” on their homepages and in some individual stories to assure users of the timeliness of the news and information. Given that updates are routinely made online, the present research intends to find out the types of content change updates made to the latest news story and how content change is different with regard to news topic and news type. Methodologically, it used real-time data collection to capture the dynamic updates of the latest news stories online. By continuously monitoring CNN.com’s latest news stories 24 hours a day, 7 days a week for 70 days and collecting actual content changes in them at 10minute intervals, the researcher is able to examine the content changes in the stories throughout their life. It is important to investigate dynamic news content changes online. The results of the study will add to the knowledge of how the Web has transformed news production and

Latest News Updates on the Web 4

consumption. Specifically, it will shed light on what updates are made in today’s 24-hour news environment and how they are influenced by news topic and type. Thus, the study contributes to mass communication theory building in the Internet age. Literature Review The information age and culture of speed Nicolas9 points out that extreme processes of social and cultural acceleration lie at the heart of the information age. Communication technology mediates everything and affects all aspects of human life. “The information age is, above all, about instant living.”(p. 21) Everfaster practices of communication have been introduced and made popular, thus speeding up culture to an unprecedented degree. Communications technologies are changing time relations. The availability of both synchronous and asynchronous communications technologies compresses the sense of time. The Internet, with its synchronous as well as asynchronous communication capability, can in different ways speed up the issues of externalized time. The evolution of online communication has led to many expectations that time’s structural influence will decrease.10 As a result, the meaning of socially constructed time becomes increasingly important. As the importance of time is radicalizing, the hunger for speed develops and is never appeased, resulting in a culture of speed,11 where information is sent in increasing amounts, ever more frequently and at ever higher speed just to attract attention. The culture of speed is one of the defining characters of the information age. Demand for and practice in immediacy The culture of speed certainly has found its way into news production and news consumption. Contemplating the effect of the Internet on news, Pavlik12 suggested that as a real-time medium, the Web will alter the production and consumption of news and information. The Web encourages media to “push” instantaneous news narratives online.13 News has thus become more fluid and is in a constant state of flux. Online stories are shorter and are built in several takes.14 They are disseminated immediately. Updates are made continuously to enable better representation of events and processes in real life,15 like realtime financial news.16

Latest News Updates on the Web 5

The practice of 24-hour news has brought about a never-ending, rolling news cycle, which, in turn, has changed the nature of public life.17 Using the Web to access news has become commonplace. One study found that nearly two thirds of the participants used online sources to get news at least some of the time.18 According to another survey,19 the majority surveyed reported that they were allowed to use their office Internet connection to catch up on the news. There is an intense interest in rapidly updated news, says Carol Perusso, president of latimes.com.20 Web users crave rapid-fire updates and a sense of continual evolution.21 Online news sites should live up to these expectations. The news sites that thrive will have to offer breaking news stories and continuously update them.22 In the early years, “the race among the most-read online news sites has turned into a competition to see which site can post wire copy the fastest.”23 Now, major sites are able to update their latest news stories several times a day, first with wire reports, then with staff-written copy. For example, in 2002, MSNBC.com frequently offered “latest developments” and breaking headlines concerning “America at War” on its website.24 “Busy saying nothing” What has all this much-touted emphasis on immediacy brought to the news consumer? In Television news, Lewis, Cushion and Thomas studied25 three 24-hour news channels and found that although they provide live news occasionally, all they provide is not much more than a choice of viewing time, allowing people to ‘‘catch up’’ on the news whenever they want it. They found that the label of “breaking news” is used more often as a device for maintaining drama than as a means of providing substantive new contents. Breaking news stories are in many cases predictable, routine and repetitive. This suggests that the question of who gets breaking news stories first is not a matter of great consequence, and will make little difference to most viewers most of the time. Research questions The Internet is as fast as television. Users often see breaking news stories online. Does Lewis and colleagues’ conclusion about 24-hour news channels apply to online news? This is a question worth pursuing. Examining the dynamic changes in the content of the latest news stories will provide an answer to the question, which, in turn, will shed light on

Latest News Updates on the Web 6

the transformation the Internet has made on the never-ending, rolling news cycle in today’s 24-hour news environment. The present research focuses on CNN.com, one of the top online news sites in the United States. It is the fastest in covering breaking news and is likely to offer a different version from that of the wires.26 It also excels at providing context and commentary. These qualities add to its popularity—when there is a surge in the demand for news, CNN.com is likely to see the biggest increase in visitors.27 Four research questions are posed about content change and how it is different concerning news topic and type. First of all, “[o]nline breaking stories are very iterative— you print what you know when you know it and then you add to it.”28 Updates are regularly made to news stories to better represent events and processes in real life. These updates result in content changes. However, not every change is prompted by the need to update the latest developments. Other considerations may also call for modifications to the posted stories. For example, when a spelling error is spotted, it warrants immediate correction, which results in a change not directly related to the event that is going on in the real world. To examine different types of changes and how they differ with respect to news topic and type, it is an important first step to examine the typology of change. Thus research question one: RQ1: What types of content change can be identified at CNN.com? Wu and Bechtel’s study of website use and news topic29 finds that there is either a positive or negative relationship between news site usage and news topics. This suggests that a site is popular with its users on some topics and unpopular on other topics. A news site cannot afford to ignore the interests of its users as it competes with other news sites for eyeballs. To attract and retain more users to its site, it is likely that more attention is given to the popular topics. Given the assumption, it is likely that a news site posts more stories on the popular topics. The more popular a topic is, the more stories it produces. In this context, the most popular topic is the one that produces the most stories. In addition to producing more stories, a popular topic may bring about more updates to the stories. Thus research question two: RQ2: How is content change different between the stories of the top three most popular news topics and those of other news topics at CNN.com?

Latest News Updates on the Web 7

Wu and Bechtel tested the relationship between news site traffic and episodic and thematic news coverage. They found that episodic coverage brings in more users to the news site than thematic coverage. The results indicate that episodic coverage captures news audiences’ attention. This preference of episodic coverage over thematic one may well be reflected in the content change of the stories. Thus research question three: RQ3: How is content change different between stories of episodic coverage and those of thematic coverage at CNN.com? For the television news industry, “live” and “breaking” news stories hold the key to increasing viewership. During breaking news events, media dependency increases.30 Similar tendency is found online. Not only do the users rush to news sites for breaking news, but they also go back frequently for new developments. Wu and Bechtel’s study31 indicates that users of online news are interested in disruptive news stories. When the dominant news story of the day is disruptive, greater web traffic brings in more users. As disruptive events are more likely to command immediate and frequent attention of the news users, news sites in these situations offer “breaking news” and “latest developments” to satiate their cravings for updated news and information, which may be reflected in the content change of the stories. Thus research question four: RQ4: How is content change different between stories of different levels of disruptiveness at CNN.com? Defining content change Content change refers to instances of addition or deletion of paragraph(s), modification of sentence(s), change of word(s), letter(s), number(s), punctuation(s) and other symbol(s). Addition or deletion of white spaces was not counted as content change. Method Unit of analysis The unit of analysis is CNN.com’s latest news story, defined as the breaking news story plus the textual story that is accessed through one of the headlines in the “latest news” column on CNN.com’s homepage. Only textual stories were examined. Audio and video clips were excluded.

Latest News Updates on the Web 8

Data collection and sampling Data collection was conducted in two steps. The first step was to collect all updates to the latest news stories at CNN.com during a period of time. To achieve this purpose, it is imperative to reiteratively collect the latest news stories from the site at regular and reasonably short intervals as well as over a relatively long period of time so that all actual updates to each story are maximally captured. In this research the sample collection was conducted once every ten minutes over the period between June 18 and August 27, 2007. This enables the researcher to obtain CNN.com’s updates to each latest news story throughout the story’s entire life. When sampling started on June 18, it was an uneventful day. However, major disruptive events happened during the period, such as an airliner crash in Brazil, a South Korean hostage crisis, a major highway bridge collapse in Minneapolis and the debate and defeat of a major immigration overhaul in the United States. Altogether, 1311 unique stories were collected, each with all its updated versions if it was updated at all. Out of the 1311 unique stories, 696 were updated at least once (meaning having at least two versions) during their lifetime as latest news stories, however minor the updates were. Further sampling was conducted to reduce the size of the sample for thorough analysis, which led to the second step. Random sampling was employed in this step. Each of the 696 stories was assigned an identification number and random numbers were generated by SPSS. The step randomly selected 228 stories, which are the sample of the present research. Coding story on news topic and type Each of the 228 sampled stories was coded for news topic of the story, whether the story was episodic or thematic, and the level of disruptiveness of the story. The coding sheet was borrowed from Wu and Bechtel’s study of Web site use and news topic and type.32 No changes were made to the coding sheet other than adding a few examples for further clarification. For example, “terrorist attack,” “hostage crisis,” “highway bridge collapse,” “coup d'état” etc. were added as additional illustrations of “extremely disruptive” news coverage. “Rescue operation after the bridge collapse” and “issues arising from the bridge collapse, such as underfunding of infrastructure” were added respectively as additional illustrations of “episodic” and “thematic” news coverage.33 The addition made the coding sheet more suited to code the news stories between June and August, 2007.

Latest News Updates on the Web 9

Two coders coded the stories. Intercoder reliability was measured by Holsti’s formula.34 It was satisfactory, with the judgment variables of topic receiving intercoder agreement of 90.15%, disruptiveness 97.52%, and episodicity 98.04%. Analysis of content change The analysis of content change was conducted on the paragraph level and between the two adjacent versions of the same story. In other words, each paragraph in the earlier version is compared with every paragraph in the subsequent update version and the comparison is made between every pair of adjacent versions. When two paragraphs are compared, one from the earlier version and the other from the subsequent update version, there are three general outcomes—they are the same paragraphs, they are completely different paragraphs, or they are paragraphs similar to each other. If they are completely different paragraphs, it means that update is made at the story level. If they are similar, it means update is made at paragraph level, or to be more exact, to the paragraph in the updated version. The first outcome suggests no content change of any sort and so these paragraphs were discarded. The second outcome indicates content change on the story level and therefore, they were sorted into two groups, cutout and addition. Cutouts include the paragraphs that appeared in the earlier version only whereas additions include the paragraphs that appeared in the subsequent update version only. The third outcome calls for close examination to classify the instances of modification, which was done in two steps. First, pairs of similar paragraphs were analyzed to establish a preliminary set of categories. The analysis was repeatedly done until no new category emerged. Then, the preliminary set of categories was applied to a random sample of 20 percent of the stories. This step was taken to discover categories that might be missed in the first sweep, thus ensuring the completeness of the categories. Since the categories of modification are limited in number, the categories were reliably established after the second sweep because no new categories emerged in later analysis of content change. The categories were then used to identify content change in the sample stories. After all the instances of content change were identified and classified for each story, they were counted and the results were entered into SPSS, where outliers were identified separately by the application on news topics, whether they were episodic or thematic and the levels of

Latest News Updates on the Web 10

disruptiveness. After all the identified outliers were removed, t-tests and ANOVA were performed with significance level set at .05, two-tailed. Results The results show that during its lifetime of nearly thirteen hours as latest news, a story is updated 3.38 times at an interval of 161.5 minutes. The characteristics of the sample are presented in Table 1, where story growth refers to the number of words in the last version of the story that are more than the first version, and story length is the average length of all versions. Table 1. Characteristics of the Sample Interval between updates Update count (minutes) Mean

3.38

161.50

Story life (minutes)

Story growth (words)

Story length (words)

763.74

99.1

672.45

N=228

Research question one asks about types of content change there are in updates. The study identified twelve mutually exclusive categories of content change, which are conversion of measurement unit, grammar alternative, punctuation alternative, spelling alternative, rewording, error correction, detailing, source identification, updating facts, change of link to multimedia, cutout, and addition. A detailed description of the categories and their examples are presented in Table 2, with cutout and addition not included because they are already explained above. Changes that fall into some of the categories significantly modify the content of a paragraph or the story as a whole. They constitute real content changes, which are labeled as substantive content change. Other changes are apparent but not real because they cause little change in the content of the story. They are quasi content changes, which are labeled as nonsubstantive content change. Categories of substantive content change include detailing, change of link to multimedia, source identification, updating facts, cutout, and addition whereas types of non-substantive content change include conversion of measurement unit, grammar alternative, punctuation alternative, spelling alternative, rewording, and error correction. Table 3 presents the distribution of content change in the categories.

Latest News Updates on the Web 11

Table 2. Categories of Content Change* Type Description Conversion of Addition or deletion of conversions between metric and the United States customary units. For example: measurement Old: “At least 75 people were wounded by … the city of Sulieman Pek, 50 miles south of Kirkuk” unit New: “At least 75 people were wounded by … the city of Sulieman Pek, 50 miles (80 kilometers) south of Kirkuk” Grammar Different use of grammatical rules that are acceptable either way. For example: alternative Old: “Mission managers: Gouge in space shuttle's belly poses no threat…” New: “Mission managers: Gouge in space shuttle's belly posed no threat…” Punctuation Different use of punctuations that are acceptable either way. For example: alternative Old: “On Saturday a sixth bore hole drilled into a chamber … the miners' families told CNN” New: “On Saturday, a sixth bore hole drilled into a chamber … the miners' families told CNN” Rewording Different choice of words or phrases. For example: Old: “Padilla was originally arrested on accusations that he planned to set off radioactive ‘dirty bombs’ ….” New: “Padilla was originally arrested and accused of plotting to set off radioactive ‘dirty bombs’ ….” Spelling Different way of spelling a word or phrase, such as abbreviation or acronym. For example: alternative Old: “… surveyed damage Friday in Rushford, Minnesota, …” New: “… surveyed damage Friday in Rushford, Minn., …” Error correction Correcting grammatical or typographic errors. For example: Old: “Snow on Thursday told hinted he might leave…” New: “Snow on Thursday hinted he might leave…” Detailing More details are provided. For example: Old: “…touches down Tuesday in Florida after its 13-day mission” New: “…touches down Tuesday at Kennedy Space Center in Florida after its 13-day mission” change of link Addition, deletion or change of a link to a video clip related to the story. For example: to multimedia Old: “Endeavour's crew members had to … on the docket for Sunday. Watch as Hurricane Dean wasn't the only worry” New: “Endeavour's crew members had to … on the docket for Sunday. Watch Endeavour touch down” Source Addition, deletion or change of the source of specific information in the story or the whole story. For example: identification Old: “Adding to the tension, … and killed at least 12 people, sources told CNN” New: “Adding to the tension, … and killed at least 12 people, sources said” Updating facts Change of specific facts that are warranted by the development of the news event. For example: Old: “BAGHDAD, Iraq (CNN) A suicide bomber … Thursday killing at least 13 people, police said” New: “BAGHDAD, Iraq (CNN) A suicide bomber … Thursday killing at least 16 people, police said” *The underlined part indicates modification.

As the present research set out to investigate how content change is different with regard to news topic and type, the following examination is focused on substantive and nonTable 3. Distribution of Content Change in Category Substantive content change Non-substantive content change

Type

cutout

detailing

addition

link

source

update

distribution

0.259

0.072

0.324

0.049

0.033

0.053

Type

conversion

grammar

punctuation

rewording

spelling

correction

distribution

0.024

0.041

0.031

0.053

0.025

0.037

substantive content change, instead of the individual categories. However, of the twelve categories identified, two are of special interest, change of link to multimedia content and error correction. They are also included in the following examination. Table 4 presents the average numbers of content change in news topic and type.

Latest News Updates on the Web 12

Table 4. Content Change and News Topic and Type Substantive content change

Non-substantive content change

Change of link to multimedia

Error Correction

Most Popular topic

22.51

1.50

1.13

.69

Other topic

13.66

1.05

.46

.70

Episodic

21.22

1.38

.99

.75

Thematic

3.05

1.00

.69

.47

Extremely disruptive

36.60

2.19

1.62

.93

Moderately disruptive

21.33

1.62

1.30

.89

Non-disruptive

6.71

.34

.34

.37

News topic

News type

N=228

Research question two examines how content change is different between the top three popular topics and other news topics. The results show that stories of the top three most popular topics have more substantive content change (t (200) = 2.823, p < .01) and more changes of link to multimedia content (t (200) = 4.007, p < .001) than stories of other topics. However, no differences are found in non-substantive content change (t (201) = 2.056, p = .041*) and error correction (t (201) = .065, p = .949). Research question three examines how content change is different between episodic news coverage and thematic news coverage. The results show that stories of episodic coverage have more substantive content change (t (194) = 4.776, p < .001) than stories of thematic coverage. However, no differences are found in non-substantive content change (t (201) = 1.418, p = .158), changes of link to multimedia content (t (207) = 1.351, p = .178) and error correction (t (82) = 1.579, p = .118). Research question four examines how content change is different between stories of different levels of disruptiveness. Differences are found in all the areas examined, which are substantive content change (F (2, 201) = 27.821, p < .001), non-substantive content change (F (2, 193) = 24.794, p < .001), changes of link to multimedia content (F (2, 208) = 17.710, p < .001), and error correction (F (2, 214) = 5.798, p < .01). Post hoc tests show that differences are found only between non-disruptive stories and stories of extreme and

*

2-tailed

Latest News Updates on the Web 13

moderate disruptiveness, except substantive content change, where differences are found between stories of each of the levels of disruptiveness. Discussion The online world is a highly competitive market environment, where established news organizations not only compete against each other, but also compete with other news services native to the Internet as well as with wire services. The competition may bring out more timely updates, resulting in content change, which better represent the events and processes in real life. The present research has produced empirical evidence of the actual content change in the latest news updates at CNN.com during the period between June and August, 2007. The findings point to three conclusions. First of all, the findings suggest that the latest news stories at CNN.com are markedly different from what is found about 24-hour news channels concerning immediacy. Whereas the news channels may just provide a choice of viewing time, allowing people to “catch up” on the news when they want to, latest news stories at CNN.com do more than that—they provide latest news and information. Timely content is a noticeable hallmark of the latest news at CNN.com. This conclusion is supported by the distribution of content change. Substantive content change accounts for 79 percent of all the changes to a story, which are updated 3.8 times and at an interval of two and half hours between updates. While an average of 3.38 updates at an interval of two and half hour may not be very impressive regarding timeliness, it must be pointed out that it is just an average. As there is considerable difference among the news topic and type, stories on the most popular topics and in popular types get much more content change much more often. The above conclusion is also supported by the findings of content change in stories of disruptive news. As one of the transcendent advantages of the Web, ease of content change makes it a preferred real-time medium for disruptive news coverage. “You print what you know when you know it and then you add to it.” CNN.com’s reporting on disruptive news demonstrates that it makes use of the advantage. They print what they know when they know it, and then take care of other aspects later—they add to it, they revise it and they correct errors. They add to it with substantive content change, multimedia included, as indicated by the big difference in substantive content change and change of link to multimedia between stories of disruptive and non-disruptive news coverage. Then they revise it, as indicated by the big difference in

Latest News Updates on the Web 14

non-substantive content change, where stories of disruptive news coverage get over 4.8 times as many polishing changes as those of non-disruptive news coverage. Then, they correct errors, as indicated by the big difference in error correction, where stories of disruptive news coverage get 2.4 times as many error corrections as those of non-disruptive news coverage. All these are clear evidence that efforts are made to ensure that timely materials are provided to the online news audience in a timely manner. In addition to timeliness of content change, the finding suggests that the updates may provide “substantial background information,”35 as indicated by the growing length of the updated story, which ends up with 99.1 words longer than its original version, representing an average of 17.3 percent growth. Again, this is an average. Stories of the most popular topics and types grow much longer. In this context, the question of when to get the latest news is a matter of consequence. For those who are interested in the news topics or in the news types and keen on the latest developments or those who rely on the news provider to help them make sense of what is happening, it is worthwhile to return to the site’s latest news stories regularly. This is the second conclusion drawn from the results. Finally, the findings indicate that users’ interest in disruptive and episode-oriented news stories are catered for in the form of substantive content change. Most of the updates are focused on disruptive and episode-oriented news stories. On average, extremely disruptive news stories get 71.6 percent more substantive content change than moderately disruptive news stories, which get 217 percent more substantive content change than nondisruptive news stories. Likewise, episodic news stories are more likely to get substantive content updates than thematic news, 21.22 compared to 3.05, representing a 5.96 times as many instances of change. All these content changes may satisfy news consumers’ craving for rapid-fire updates and create a sense of continual evolution. A limitation of the study is that the update results should be interpreted as an approximation. Both reporting breaking news and updating posted news stories are rolling processes. They can occur at any moment. An artificially-set regular data collection of once every ten minutes may fail to capture every single instance of content change to the latest news story during its lifetime. Due to the limitation, the count of content change actually indicates the lower bound, meaning that, on average, instances of content change made to a story are at least as many as the number indicates. Methodologically, it is a tradeoff between

Latest News Updates on the Web 15

the accuracy of approximation and the need for minimizing potential disruptions in data collection to ensure the quality of the data collected. Setting shorter data collection intervals, such as two or five minutes could make the approximation more accurate but it is still an approximation. However, shorter intervals may cause more disruptions in data collection due to technical problems, such as network outage, dramatically increasing the rate of unusable data. Conclusion The Internet has brought about a culture of speed, where news production has undergone transitions and transformations. By collecting dynamic real-time updates in the latest stories at CNN.com 24/7 for 70 days and content analyzing the changes in the updates, this research has produced important empirical findings about the online news updates with regard to news topics and news types. The knowledge of online news updates contributes to the development of mass communication theory, and this research is just another starting point for more research, both exploratory and explanatory, into the news on the Web.

Latest News Updates on the Web 16

1 2 3 4

5

6 7 8 9

10 11

12

13

14

15 16 17

18 19

20 21

22 23 24 25

26 27 28 29

30

31 32 33

J. Vince, “When the media meet as one.” UNESCO Courier, 54 no. 10, (2001): 44-45. Palser Barb “Brave old world,” American Journalism Review, 23, no. 6 (2000): 74. Ibid. Philip Seib, Going live: Getting the news right in a real-time, online world (Lanham, MD: Roman & Littlefield Publishers, 2001). Nielsen/Netratings, “Traffic to daily newspaper and finance news sites jump, according to Nielsen/Netratings.” (2002) A. Colon, “The multimedia newsroom.” Columbia Journalism Review, 39, no. 1 (2000). Barb Palser, “Not quite ready for prime time.” John V. Pavlik, Journalism and new media. (NY: Columbia University Press, 2001). Gane Nicholas, “Speed up or low down? Social theory in the information age.” Information, Communication & Society, 9 no. 1, (2006): 20-38. Gotved Stine, “Time and space in cyber social reality,” New Media & Society, 8 no. 3, (2006): 467-486. Jan van Dijk, (2006). The network society: Social aspects of new media. London, England: Sage Publications. John V. Pavlik, “New media and news: implications for the future of journalism,” New Media & Society, 1, no. 1, (1999): 54-59. Elisia L. Cohen, “Online journalism as market-driven journalism,” Journal of Broadcasting & Electronic Media, 16, no. 4, (2002): 532-548. Karla Aronson and George Sylvie, “Real-time journalism: Implications for news writing,” Newspaper Research Journal, 17 no. 3/4 (Summer/Fall96). John V. Pavlik, Journalism and new media. Karla Aronson and George Sylvie, “Real-time journalism: Implications for news writing.” Justin Lewis, Stephen Cushion and James Thomas, “Immediacy, Convenience or Engagement? An analysis of 24-hour news channels in the UK,” Journalism Studies, 6 no. 4, (2005): 461-477. Donna Shaw, “Online scoop,” American Journalism Review, 28, no. 5, (2006): 58-60. Swartz, Nikki, “Employees addicted to online news,” The Information Management Journal, 36 no. 6, (2002): 8. Kelly Heyboer, “Going live,” American Journalism Review, (January/February 2000). M. J. Pitts, “Television web sites and changes in the nature of storytelling,” Studies in Media & Information Literacy Education, 3, no. 3, (2003). Kelly Heyboer, “Going live.” As cited in Elisia L. Cohen, “Online journalism as market-driven journalism.” Elisia L. Cohen, “Online journalism as market-driven journalism.” Justin Lewis, Stephen Cushion and James Thomas, “Immediacy, Convenience or Engagement? An analysis of 24-hour news channels in the UK.” Elisia L. Cohen, “Online journalism as market-driven journalism.” Palser Barb, “Online Advances,” American Journalism Review As cited in Donna Shaw, “Online scoop.” Denis H. Wu and Arati Bechtel, “Web site use and news topic and type,” Journalism & Mass Communication Quarterly, 79, no. 1 (2002): 73-86. Andrea Miller, “Watching viewers watch TV: Processing live, breaking, and emotional news in a naturalistic setting,” Journalism & Mass Communication Quarterly, 83, no. 3, (2006): 511-529. Denis H. Wu and Arati Bechtel, “Web site use and news topic and type.” Denis H. Wu and Arati Bechtel, “Web site use and news topic and type.”

The coding instruction and sheet are available upon request from the author.

Latest News Updates on the Web 17

34

Wimmer R., & Dominick J. (2000). Mass Media Research. Belmont, CA: Wadsworth, 151-152. Reliability = 2M/(N1+N2), where M is the number of coding decisions on which two coders agree, while N1 and N2 are the total number of coding decisions by the first and second coder, respectively. 35 Denis H. Wu and Arati Bechtel, “Web site use and news topic and type,” pp. 81-82