News Value of Sports Coverage:
A Test of Newsworthiness Model on the World Cup Coverage
by
Yun Jung Choi (Master's student)
Jong Hyuk Lee (Master's student)
Cheolhan Lee (Doctoral Student)
University of Missouri at Columbia
School of Journalism
2607 Eastwood Dr. # 63
Columbia, MO 65202
(573) 474- 1495
[log in to unmask]
*** Student Paper
A paper submitted to Communication Theory and Methodology Division
AEJMC 2003 Annual Conference
Kansas City, Missouri
July 20 August 2, 2003
The News Value of Sports Coverage:
A test of Newsworthiness Model on the World Cup coverage
Abstract
The purpose of this study is to test the international newsworthiness model
of Shoemaker, Danielian, and Brendlinger (1991) in sports coverage. The
World Cup held in Korea and Japan was selected as the topic. To test the
hypothesis, significance and deviance points of the World Cup matches were
operationally defined based on the FIFA ranking points, William Hill
betting points, and CNN power ranking, and then these statistics were
compared with the coverage.
A regression analysis showed that the U.S. media, the Korean media, and the
U.S. online media were generally influenced by the prediction of the
newsworthiness model. The more significant a match was, the more
prominently the three media covered it. The more deviant a match was, the
more prominently it was covered by the media. There were also mixed
effects. If a match were both significant and deviant, more attention was
given to the match than matches that are only significant or only deviant.
Overall, the world newsworthiness model was shown to be efficient in
predicting the amount of sports coverage. However, the newsworthiness model
failed to predict the coverage of the Korean online media. The study also
suggests that the newsworthiness model was a better predictor of the
traditional media than the online media.
Introduction
The 2002 World Cup soccer games began with a surprise. France, the
defending champion of the 1998 World Cup and the top FIFA (Federation de
International Football Association) ranked team, was stunningly defeated by
their former colony, Senegal. The next day, the headline that adorned the
first page of New York Times sports section read "In a Stunner, Senegal
Tops The Champion" (New York Times, 2002, June 1, Section D, p.1). During
the 16 days of 64 preliminary matches that followed the opening match, an
unusual number of upsets and surprises brought people's attention to the
World Cup games.
These upsets provide an intriguing opportunity to examine the news values
of sports coverage. How would the media react differently to outcomes that
are expected and unexpected? What attributes of sports events would draw
more attention from the media?
Studies of news values have been conducted by many scholars for several
decades. The accumulated research has already established several criteria
that journalists depend on in selecting news: timeliness, proximity,
novelty, impact, human interest, and unusualness. (Galtung & Ruge, 1965;
Rosenberg, 1970; Ostagaard, 1965). Shoemaker, Danielian and Brendlinger
(1991) suggested two important news value determinants significance and
deviance in their newsworthiness model. They concluded that the more
significant and deviant an event is, the more coverage the event receives.
The present study, then, tries to examine whether this model of
newsworthiness applies to sports news coverage.
The World Cup event held in Korea and Japan in June 2002 was chosen as the
topic. The design involved measuring the degree of significance and
deviance of each match, and then examining the matches that received the
most coverage. Raw data about the competency of each team such as FIFA
ranking or international betting odds indexed the degree of significance
and deviance of each matches. If the first ranked team and second ranked
team had a match, the match would be regarded as more significant than
other matches in which low-ranked teams participated. If the first-ranked
team was defeated by a low-ranked team, as was the case in the
France-Senegal game, the match would be considered more deviant than other
matches where a low-ranked team was beaten by a top-ranked team. With this
logic, every match will be measured in the degree of significance and
deviance.
Based on the measured data, this study will find out if there was any
relationship between measured news values and media coverage. The study
will be conducted in four different contexts: the U.S. media, the Korean
media, the U.S. online media, and the Korean online media. This study would
contribute to the traditional research on news values in several ways.
First, the discussion of news values would be tested on sports news
coverage beyond political, economic and social news. Second, this study
could expand news value research to online news coverage. Third, this study
tests the model in two highly different cultures, the U.S. and Korea.
Literature Review
News Values
Many have attempted to define news, but none have provided an absolute
answer. Stewart Hall (1981) has written "news value is one of the most
opaque structures of meaning in modern society. All true journalists are
supposed to possess it; few can or are willing to identify and define
it"(p.234).
Park (1967) regarded news as a special form of knowledge that has certain
implications for human behaviors and mobility. Stephen's definition was
"new information about a subject of some public interest that is shared
with some portion of the public." (Stephens, 1988, p. 9.) Breed (1956)
proposed the following definition: "News is the report of a recent event
judged by newsmen to be worthy of publication for the interest and /or
information of members of their audience, and has the following
characteristics to a greater or less extent: recency, interestingness,
accuracy, availability, simplicity, saleability, significance, prudence,
objectivity and superficiality; it is frequently mediated by an
association, increasingly interpretive and is journalistically stylized in
form"(p.447).
One way to define "news" is to study news values. Many scholars have
accumulated a vast amount of literature on newsworthiness. Most news value
studies has expanded international news. Galtung and Ruge (1965) suggested
that events tend to become news if they have 12 factors: frequency,
threshold, intensity, unambiguity, meaningfulness, consonance,
predictability, unexpectedness, continuity, composition and relevance to
elite nations, elite people or something negative. Rosenberg (1970) added
three concepts in the study of international news: degree of importance of
the event, physical or cultural distance of the events, and degree of the
predictability of the events. Hester (1973) approached the study of
international news and information flow from the international relations
perspective. He proposed a hierarchy of nations, cultural affinities and
economic association between nations, and news and information conflicts as
international news determinants. Ostgaard (1965) paid attention to the
factors influencing the flow of news in terms of political and economic
systems: simplication of the news, identification with the news,
sensationalism, government control of the mass media, and economic
consideration such as media ownership. Gans (1979) included social values
in factors influencing international event coverage. He proposed
ethnocentrism, altruistic democracy, responsible capitalism, moderatism,
social order, and national leadership as factors that influence journalists
when deciding which international news should get covered or be given more
space or time.
One paradigm that explains world news flow is the world system theory
(Wallerstein, 1974), which posits that the global system is fundamentally
composed of a two-tier concentric market, with the Western world in the
center and the rest of the world in periphery (Santons, 1996; So, 1990).
Using network analysis, Kim and Barnett (1996) contended that there is
"inequity in international news flow between the core and the periphery.
The Western industrialized countries are at the center, dominating the
international news flow. Most African, Asian, Latin American, and Oceanian
countries are at the periphery." The core country in the world system is
more likely to occupy the central position in the international
communication network, whereas peripheral countries are more likely to be
influenced by the core nation's media. Against the backdrop of the world
system perspective, Chang (1998) tried to draw an international news system
with analysis of Reuter coverage of the WTO (World Trade Organization)
conference. The finding suggests that core nations have higher chances to
be in the news than those countries located in the semi-periphery or
periphery. In contrast, nations in the peripheral zones have to go through
several filters before they make it to the news.
Among various attempts to identify the news value factors, the
newsworthiness model <figure1> suggested by Shoemaker, Danielian, and
Brebdlinger (1991) is considered to include a comprehensive range of
factors that affect international news values under two broad concepts
deviance and social significance. Deviance concepts include novelty,
oddity, unusualness, prominence, sensationalism, and conflict or
controversy. The significance dimension consists of importance, impact,
consequence, and interest. Social significance refers to how significant a
foreign event is to the U.S. This concepts is further sub-divided into (1)
political, economic, and cultural significance, (2) communication
constraints, and (3) event prominence in the U.S. media. In addition,
Shoemaker and her colleagues added two more concepts proximity and
timeliness as sub-categories under the deviance dimension.
Shoemaker and her colleagues tested the newsworthiness model and found that
the New York Times and three major TV networks in the United States tend to
cover international events prominently according to their level of
deviance. In addition, they found political and economic significance was
positively related to both newspaper and television coverage. They also
found that social significance correlated occasionally with the deviance
and proximity dimensions. According to the newsworthiness model, the more
socially significant and deviant an event is, the more coverage it gets.
News Values Across Cultures
The newsworthiness model of Shoemaker, et al. (1991) provides efficient
criteria to predict and estimate news values of an event because the model
includes a broad range of news value factors. However, the model cannot be
assumed to be applicable to every country in the same way because every
country has different politics, social structure, cultural characteristics,
and physical constraints. Also journalists in different countries might
have different news values and agendas depending on their society's
ideology and their country's position in the global system. Shim (2002)
pointed out that constraints in news making processes are different in each
country, and audiences demand different information depending on the
society they reside in.
Although not much attention has been paid to the newsworthiness models of
third world or peripheral countries, different news values according to
different countries have been implicated by studies on international news
flow. The vast amounts of news value studies are concentrated on how the
U.S. media cover international events, but not the other way around.
One of the empirical analyses on different news coverage in diverse
countries was conducted by Cooper-Chen (1996). The study, comparing 309
stories selected from Tokyo Broadcasting System newscasts (Japan) with 283
stories from CBS newscasts, concluded that Japan allowed less room for the
outside world than the U.S. media. In addition, geographic areas covered
were significantly different in two countries.
Wu's study (2000) proposed factors that influence international news
coverage, which are traits of nations, relatedness between nations, and the
logistics of newsgathering. The study compared the international news
coverage of 38 countries and concluded that several factors differentiate
the amount of coverage of international events in different countries.
Among the 38 countries, the U.S. was the most covered country in the
world. Robinson and Sparkes (1976) investigated the international news
coverage of 39 newspapers in Canada and the U.S. and discovered that the
two countries do not apply the same criteria in selecting news stories.
Trade was found to be correlated with the coverage of Canadian newspapers,
whereas it did not correlate with the U.S. media.
Many studies have supported the hypothesis that colonial ties influence
international news coverage. Nnaemeka and Richstad (1980) surveyed 19
newspapers in the Pacific region and found that the nations having colonial
ties will be given more attention within the same colonial group. The
studies of Atwood (1985), Meyer (1989), and Shurnik (1981) echoed Nnaemeka
and Richstad's finding that old colonial ties were also found to be an
important factor in determining the volume of the news flow.
Contrary to other findings, Shim (2002) applied Shoemaker's model of
newsworthiness to a new industrialized country South Korea and
concluded that the model can be also used in predicting the international
coverage patterns of South Korean newspapers. The U.S. model was useful in
explaining Korean international news coverage. A content analysis of 1,229
randomly chosen articles from Korean newspapers showed that Korean
newspapers evaluated sensationalism, human interest, and structural
conflict highly when they covered international news, but gave less
attention to international news if it deviated from their social norms and
cultural values. Overall, the study concluded that Shoemaker and her
colleagues' theoretical framework of deviance and social significance is
useful and applicable to newly industrialized countries like South
Korea.
News Values of Online Media
Ever since online media were introduced, many researchers have tried to
define the online media by comparing them with traditional media. Most
critics charged online newspapers for reproducing the same print product at
online. Regan (1995) argued "simply sticking your content or shovel
ware on a web site just doesn't cut it any more" (p. 78). Although online
newspapers are beginning to provide an increasing amount of news content,
most online news sites are criticized as containing "shovel ware"
content that has been created for print media and has simply been shoveled
on the Internet. Gubman and Greer (1997) studied whether this criticism is
justifiable. They found no significant difference between traditional and
online newspapers on news content, writing styles and presentation.
However, they found that large newspaper sites were significantly more
likely to carry national news than mid-sized or small newspaper sites.
One study compared the gate-keeping process of online media with printed
media. Singer (1999) tried to explore whether online newspaper gatekeepers
provide readers with the same information printed in traditional media or
provide different media content. A content analysis of 4,786 online news
stories found that although the online media draw most of their content
from their print versions, they do not copy everything. The online media
are inclined to include local news, and the contents they leave out are
likely to be non-local news. She concluded that online newspapers'
gate-keeping role is different from the role of traditional media.
Hypotheses
The purpose of this study is to apply the newsworthiness model of
Shoemaker, Danielian and Brendlinger (1991) to sports coverage. This study
examined if the newsworthiness model was successful in predicting the media
coverage of the 2002 World Cup matches. This analysis was conducted to four
different media types U.S. traditional media, Korean traditional media,
U.S. online media, and Korean online media.
U.S. Traditional Media
H1a (significance): The more significant a match was, the more prominently
it was covered in the U.S. traditional media.
H1b (deviance): The more deviant the result of a match was, the more
prominently it was covered in the U.S. traditional media.
H1c (mixed effect): If a match was both significant and deviant, it was
covered more prominently than matches that were only significant or only
deviant in the U.S. traditional media.
Many scholars have tried to apply the international news structure and
processing in a global context. Previous research has demonstrated that
different news values exist in different countries. However, Shim (2002)
applied Shoemaker's newsworthiness model, which had been used for the U.S.
media, to South Korean media, and found that Korean media have similar news
value criteria.
Korean Traditional Media
H2a (significance): The more significant a match was, the more prominently
it was covered in the Korean traditional media.
H2b (deviance): The more deviant the result of a match was, the more
prominently it was covered in the Korean traditional media.
H2c (mixed effect): If a match was both significant and deviant, it was
covered more prominently than matches that were only significant or only
deviant in the Korean traditional media.
It is difficult to conduct news values studies of online media because most
of print news stories are replicated in online media. Singer (1999) pointed
out that out of 4,786 online newspaper articles she content analyzed for
her study, none of the articles were solely produced for online newspapers.
However, by observing what kinds of articles from the print media are
selected for online editions, news values of online media can be
measured. This paper will test if the Shoemaker's newsworthiness model can
be applied to online media. First, the model would be tested on the U.S.
online media. Based on previous researches on the online media, the
following hypotheses are tested:
U.S. Online Media
H3a (significance): The more significant a match was, the more prominently
it was covered in the U.S. online media.
H3b (deviance): The more deviant the result of a match was, the more
prominently it was covered in the U.S. online media.
H3c (mixed effect): If a match was both significant and deviant, it was
covered more prominently than matches that were only significant or only
deviant in the U.S. online media.
Korean Online Media
H4a (significance): The more significant a match was, the more prominently
it was covered in the Korean online media.
H4b (deviance): The more deviant the result of a match was, the more
prominently it was covered in the Korean online media.
H4c (mixed effect): If a match was both significant and deviant, it was
covered more prominently than matches that were only significant or only
deviant in the Korean online media.
Methodology
World Cup 2002
Methodologically, a global event such as the World Cup matches represents a
specific event that would facilitate an interesting data set to use in
examining cross-cultural news values. All countries involved potentially
have equal chances to be covered.
A total of 48 matches in the first round were selected for this analysis
out of the 64 matches in the World Cup. The rest of the matches were held
in the second round through the final game. The semifinal matches would be
more important than matches in the first round. For this reason, only the
48 matches of the first round were selected based on the assumption that
the game structure would not influence the significance of the matches.
Significance
Significance of sports events would seem to be more independent from
political, economic and political context than other international events.
For instance, Brazil and Argentina are not very related to U.S. politically
and economically, but they have been regarded as the most significant
soccer teams in the world. In sports events, significance is defined as the
competency of the team. Therefore, this study will measure the significance
and deviance of matches by using indices of international soccer teams such
as FIFA points, William Hill betting points, and the CNN power ranking.
The FIFA ranking is the most authoritative resource that predicts the
competence and ability of soccer teams. The ranking is produced by a
computer program, which assigns a team points for every international A
match result over the last eight years. For instance, France has 802 and
China has 599 FIFA points.
William Hill is one of oldest and largest online sports betting site, and
is based in the U.K. During the World Cup games, many people made bets on
the World Cup matches. Those games that were significant drew more betters
and betting money than those games that were insignificant. The William
Hill betting points were transformed because the gap between strong teams
and weak teams is extremely large. For instance, the winning chance for the
France team was 4:1 whereas the chance for the China team was only 750:1.
For this reason, William Hill betting points were transformed into a
reverse rank order. For example, England's William Hill points, 9:1, was
the 6th highest score among 32 teams, so England was given 27 points (33-6)
according to the reverse rank order design.
CNN ranked 32 qualifying teams according to their previous World Cup
performance, weak points, and key players. The CNN power ranking points
were also transformed into reverse rank order. England's CNN points, which
were the 5th among 32 nations, were recorded as 28 (33-5). <Table 1> lists
the significance points for 32 participating teams.
[Insert Table 1 here]
With three types of significance points for each team, the researchers
measured the significance points of every match. Matches between two strong
teams are considered as significant compared with other matches between a
strong team and a weak team or between weak teams. The more significant two
participating teams were, the more significant the match was. Based on this
logic, the significance of a match was operationally defined as the sum of
the significance points of the two participant teams. For instance, England
vs. Argentina will get 1478 (694+784) FIFA points, 59 (32+27) William Hill
points, and 60 (32+28) CNN points. On the other hand, the significance
points for China vs. Costa Rica were 1209 (566+643) FIFA points, 9 (2+7)
William Hill points, and 6 (1+5) CNN points. In this way, every match had
its own significance points in three categories of the FIFA, the William
Hill, and the CNN points. <Table 2> lists significance points for 48 matches.
[Insert Table 2 here]
Deviance
The concept of deviance for sports events was expected to relate to upsets.
If a weak team defeated a strong team, the event was deviant because the
result is against people's expectation. The deviance of sports events,
therefore, would be defined as upsets. Wanta and Leggett (1988) also
compared expected outcomes with actual outcomes to produce a variable
called "degree of upset." A favored team losing or winning by less than the
pre-game betting line was conceptualized as "upsetting" in their study.
To measure the deviance points of every match, the present analysis broke
down every match into three groups: upset, draw, and expected outcome. The
upset group included all matches where weak teams with lower significant
points defeated teams with higher significant points. France vs. Senegal
belonged to this group because the lower-ranked Senegal team (599 FIFA
points, 10 William Hill points, and 7 CNN points) defeated France team with
higher significance points (802 FIFA points, 31 William Hill points, and 32
CNN points). The draw group included every match whose result turned out to
be a draw. The expected outcome group included non-deviant matches where
favored teams defeated expected losers. Among the three groups, the degree
of deviance was higher in the order of the upset, the draw, and expected
outcome groups.
The next step was to rank every match in terms of the degree of deviance
within the group, with the consideration of the significance point
difference between the two teams. In the upset and draw group, the larger
the significance point difference between two teams, the more deviant the
match. Even though Croatia (655 FIFA points) beat Italy (717 FIFA points),
the match would be perceived as less deviant than France (802 FIFA points)
beating Senegal (599 FIFA points).
In the expected outcome group, however, the measuring logic should be used
differently. In this case, the wider the significance point gap between
two teams was, the less deviant the match could be perceived. The match
where Brazil defeated China would be regarded less deviant than the match
in which Brazil defeated Turkey, even though the two matches ended with
expected results. The significance point gap of the former match (218 FIFA
points) was wider than that of the latter match (130 FIFA points). In this
way, every match in the expected outcome group was given its rank below the
upset and the draw group.
Based on this measurement, every match had its own deviance points in rank
order. For easy comparison with the media coverage, the deviance rank was
transformed into a reverse rank order. As the result, France vs. Senegal
was given 48 points as the most deviant match, and Germany vs. China was
given 1 based on FIFA points. <Table 2> lists ranked deviance points for
the 48 matches.
Sampling
After the 48 matches were assigned significance and deviance points, it was
necessary to examine how each match was covered by news media. Eight news
media were selected: the New York Times and the Washington Post (the U.S.
media), the online editions of the New York Times and ESPN (the U.S. online
media), Chosun-Ilbo and KBS (the Korean media), and online editions of
Chosun-Ilbo and KBS (the Korean online media).
The New York Times and the Washington Post have been regarded as the most
influential media in the U.S. ESPN was the main broadcasting media that
covered matches exclusively. As for the Korean media, Chosun-Ilbo is the
most circulated daily newspaper, and the KBS (Korean Broadcasting System)
is the biggest public broadcasting company in Korea.
The sampling period was 16 days ranging from May 31st to June 15th, when
the first round of the World Cup games took place. For the New York Times
and the Washington Post, researchers searched Lexus/Nexus with a keyword of
"World Cup." The Chosun-Ilbo and KBS news stories were obtained by
searching KINDS (Korean Integrated News Database System) with the same
keyword. The population for the Chosun Ilbo was the whole newspaper
articles, and the population for the KBS was the "KBS 9 O'clock News" which
is the main news program.
Regarding online versions of news media, researchers downloaded the World
Cup special pages of online media once a day during the 16 days of the
World Cup games. The special World Cup pages of the online media were
captured and printed at noon Central Standard Time in the U.S. This time
was when all the game results came out, deadlines for both the U.S. and
Korean media are past, and the online versions were updated.
Articles and broadcasting news stories that directly related to the World
Cup games, players, and coaches were selected for the study. News stories
that looked at World Cup "fever," enthusiastic fans and the popularity of
soccer players were eliminated from the sample. Only the stories that
reported on matches, such as match results, match strategy, performance of
players during the game, leadership of coaches, and analysis of games, were
included in the sample.
For the U.S. media samples, the articles about games that included the
U.S. team were excluded in order to control contaminating factors that
could influence the significance and deviance of news values. By
eliminating the U.S. matches that are obviously most importantly reported
by the U.S. media, other matches have equal significance and deviance. For
the same reason Korean matches were excluded from the Korean media
sampling. Through this process, the eight news media produced a total
number of 485 stories <Table 3>.
[Insert Table 3 here]
Coding Unit
Every selected article was analyzed to see how prominently it covered each
match. It was not possible to code an entire article into a match because
many articles included details of more than one match. The coding unit,
therefore, was defined as a "match" mentioned in three or more sentences in
an article. A total of 592 matches were counted as coding units <Table 3>
Category and Importance Coverage Point
Every match was coded into several categories: placement, headline,
placement within the article, and length.
Placement "Placement" means on what page of newspaper the match was
covered. A match that was covered on the front page was given 4 points, and
a match that appeared on the rest of general section pages was given 3
points. A match covered on the first page of the sports section was given 2
points, and stories on the rest of the sports section pages received 1
point. In case of KBS, a match aired within the first five stories of the
whole news program was coded as 4 points. Three points were given to
matches within 10th story, two points were allocated to matches within
15th, and the rest of matches in the news were scored one point.
Online versions of media mostly had 3 to 8 main news stories and 5 to 10
sub-stories. The stories posted at the most upper side of the screen were
given 4 points, the second story was coded as three points, the third story
was given two points, and the rest of stories received one point each.
Headline The headline category is fairly obvious. Researchers examined
whether the name of teams or players involved in a match were mentioned in
the headline. If so, the match obtained 1 point, and if not, it was given 0
points.
Placement Within an Article Many articles listed several matches in one
article. Therefore placement within an article had to be measured. The
placement within an article indicates at which part of the article the
match news was located. If located in the upper part of an article, the
match news received one point, and if located at the lower side, the match
news was given 0 points. However, this category was not applied to the
online media because downloaded World Cup special pages did not display
every sentence of articles.
Length The length category was established based on the number of words.
Regarding the New York Times and the Washington Post, matches with more
than 750 words was coded as 4 points, matches with more than 500 words was
given 3 points, and matches with more than 250 words received 2 points. The
matches that were summarized in less than 250 words were coded 1 point. The
length categories were determined based on a preliminary examination of the
length of 20 % of selected articles so that all news could be divided into
four groups almost evenly. With the same preliminary preparation, the
Chosun-Ilbo had four groups in the length category: more than 1000 words (4
point), 701 to 1000 (3 point), 401 to 700 (2 point), and 400 or less (1
point). In case of KBS, matches with more than 600 words were coded as 4
points, matches with 501 to 600 words were coded as 3 points. Matches with
301 to 500 words were given 2, and rest of matches received 1 points. Again
the length category was not applied to the online media.
In this way, every match was coded into every category with points. With
this design, every match received the "Importance Point", defined as the
sum of every point in every category. The Importance Point indicates how
prominently the match was covered by media. The maximum could be 10
(placement 4 points + headline 1 points + placement within an article 1
points + length 4 points) and the minimum could be 0. The Importance Point
for the 48 matches are summarized in <Table 4>.
[Insert Table 4 here]
Reliability
Two different coders conducted reliability checks. Because the
coding unit was defined a "match" rather than an article, first reliability
check was conducted to identify selected matches from an article. 73
articles (15%) was randomly selected and re-coded by two coders to see if
they identified same matches from the same articles. The reliability was
89.1% according to the Holsti's formula. Then, also the 15% of matches were
randomly selected from eight media respectively to check inter-coder
reliability. The reliability was 94.5% for placement category, 97.8% for
headline, 88.4% for placement within an article, and 81.4% for length.
Results
The first set of hypotheses tested the link between significance, deviance,
the combination of the two, and the "coverage importance" scores for each
medium. The regression analysis was adopted to examine causal relationships
between media coverage and newsworthiness that were represented in
significance points, deviance points, and the combination of the two. As
shown in Table 5, significance, deviance and the mixed effects were overall
significant. Only two non-significant results appeared in The New York
Times' significance categories. All hypotheses were supported.
[Insert Table 5 here]
The second set of hypotheses dealt with the Korean media. <Table 6> shows
that significance and mixed effects were significant in both Chosun-Ilbo
and KBS coverage. However, Deviance effects were significant in only KBS
coverage. H2a and H2c were supported but H2b was partly supported.
[Insert Table 6 here]
The third set of hypotheses asked about the U.S. online media. The online
edition of New York Times showed significant results in significance and
mixed effects, but non-significant results in every deviance categories.
The online ESPN showed significant results in the mixed effect, but didn't
show many significant results in the other two effect categories. In
general, H3c was supported while H3a and H3b were partly supported <Table 7>.
[Insert Table 7 here]
The fourth set of hypotheses was about the online Korean media. The online
edition of Chosun-Ilbo showed significant results in significance and mixed
effects categories, but non-significant results in every deviance
categories. The online KBS didn't show any significant results across all
categories. H4a and H4c were partly supported while H4b was not supported
<Table 8>.
[Insert Table 8 here]
Discussion
The results shown here are important in a number of ways. First,
Shoemaker's world newsworthiness model was, in general, successful in
predicting the coverage of sports events. Significance and deviance each
predicted news coverage of matches quite well.
Second, this study found that there was not much difference in
significance, deviance, and mixed effects to influence media coverage
between the U.S. and Korean media. Both media reported significant and
deviant matches prominently. This finding supports the result of Shim's
(2002) previous finding that Shoemaker and her colleague's theoretical
framework of deviance and social significance is useful and applicable to
newly industrialized countries such as South Korea. The result implies that
Shoemaker's theory of deviance and social significance could apply
cross-culturally and can identify the current developmental status of
international reporting of other countries. The result also supports Chang,
Shoemaker, and Brendlinger's (1987) categorization of the international
news coverage into event-oriented and context-oriented approaches. The
study indicates that in an event-oriented approach where factors inherent
in the event determine its prominence regardless of the context,
international news value determinants are similar cross-culturally. World
Cup coverage is an international event that could be reported at the
event-oriented dimension. This could also imply that the context-oriented
dimensions of the news value, such as political and economic contexts did
not influence the World Cup match reporting.
Third, the newsworthiness model was more strongly supported in traditional
media than on online media. This embodies some important implications. The
traditional media such as newspapers and broadcasting news have limited
space and time to present news and have to update the news every day.
Therefore, it could be assumed that more strict criteria of news values are
applied to traditional media outlets. On the other hand, online media have
unlimited space to present news and are not influenced by the time
constraints. Some news stories on online media are updated hourly whereas
some important online news stays on for a week. For example, ESPN.com and
the New York Times online kept some important news stories for several days
while updating other stories daily. It could be assumed that the difference
of time constraints and limitation of presenting space between traditional
media and online media resulted in applying different news values. More
studies should be conducted in this area.
Fourth, mixed effects of significance and deviance combined showed more
significant results than only significance effect or only deviance effect.
In other words, matches that were both significant and deviant were more
prominently covered by the media than matches that were only significant or
only deviant. The results also revealed that the significance effect
influenced media coverage more strongly than the deviance effect.
Fifth, this study tested three indexes of significance the FIFA points,
the William Hill betting point, and the CNN power ranking in sports
coverage. Among the three significance indexes that were used, the FIFA
significance points were correlated most strongly with the coverage. FIFA
points significance were better predictor of news coverage than the William
Hill betting points and CNN power ranking. The FIFA ranking is the most
authoritative and comprehensive resource that measure the competence of a
soccer team based on multiple criteria, such as winning and losing, number
of goals, strength of opponents, regional strength, and importance of the
match. The FIFA points are produced by a more systematic analysis than
William Hill betting points which are based on participants' instincts or
the CNN power ranking, which is based on a less comprehensive analysis.
Conclusion
The purpose of this study is to test Shoemaker, Danielian, and
Brendlinger's (1991) world newsworthiness model in four contexts the U.S.
media, the Korean media, the U.S. online media, and the Korean online
media. In order to test hypotheses, significance and deviance points of the
World Cup matches were created and correlated with the coverage. The result
revealed some important findings.
The finding revealed that the U.S. media, the Korean media, and the U.S.
online media were generally found to correlate with the prediction of the
newsworthiness model. The more significant that a match was, the more
prominently it was covered by the three media. The more deviant that a
match was, the more prominently it was covered by the three media. There
were also mixed effects. If a match was both significant and deviant, more
attention was given to that match than only significant or only deviant
matches. Overall, the world newsworthiness model was efficient in
predicting the coverage of an event by the three media. However, the
newsworthiness model failed to predict the coverage of the Korean online media.
More studies should be conducted on news values of world events coverage
and online media. Further studies on more longitudinal or comprehensive
events should be conducted in world sports event coverage, such as the
Olympic games. Also more detailed analyses of news values of the online
media should be conducted. The different nature of the online media from
traditional media is not expansively discussed yet. The distinction between
online media and traditional media could be explained by finding different
new values for both media.
Reference
Atwood, L. E. (1985). Old colonial ties and news coverage of Africa. Paper
presented to the International Communication Association annual convention,
Honolulu, HI, U.S.
Breed, W. (1956). Analyzing news: Some questions for research. Journalism
Quarterly, 33, 467-477.
Chang, T. K. (1998). All countries not created equal to be news: World
system and international communication. Communication Research, 25, 528-563.
Chang, T. K., Shoemaker, P. J., & Brendlinger, N. (1987). Determinants of
international news coverage in the U.S. media. Communication Research, 14,
(4), 396-414.
Cooper-Chen, A. (1996). Two pacific powers view the world: News on CBS and
TBS television, 1993. Paper presented at the annual convention of the
Association for Education in Journalism and Mass Communication.
Galtung, J., & Ruge, M. H. (1965). The structure of foreign news. Journal
of Peace Research, 2, 64-91.
Gans, H. J. (1979). Deciding what is news: A study of CBS Evening News, NBC
Nightly News, Newsweek and Time. New York: Panthom Books.
Gubman , J. & Greer, J.(1997). An analysis of online sites produced by
U.S. newspapers: Are the critics right? A paper presented at the annual
convention of the Association for Education in Journalism and Mass
Communication held 1997.
Hall, S. (1981). The determinants of photographs. In S. Cohen and J. Young
(Eds.), The Manufacturing of news: A reader (p. 226-243). Beverly Hills,
CA: Sage.
Hester, A. (1973). Theoretical consideration in predicting volume and
direction of international information flow. Gazette, 19, 239-247.
Kim, K. & Barnett. G. A. (1996). The determinants of international news
flow: A network analysis. Communication Research, 23, 323-352.
Meyer, W. H. (1989). Global news flow: Dependency and neo-imperialism.
Comparative Political Studies, 22 (3), 243-264.
Nnaemeka, T., & Richstad, J. (1980). Structured relations and foreign news
flow in the Pacific region. Gazette, 26, 235-258.
Ostgaard, E. (1965). Factors influencing the flow of news. Journal of Peace
Research, 2, 39-63.
Park, R. (1967) News as a From of Knowledge. In R. H. Turner (Eds.), On
social control and collective behavior (pp. 32-52). Chicago: Chicago
University Press.
Regan, T. (1995). New alone is not enough. (Using the Internet to publish
news). Nieman Reports, 49 (winter), p. 78.
Robinson, G, J., & Sparkes, V. M. (1976). International news in the
Canadian and American press: A comparative news flow study. Gazette, 22,
203-218.
Rosenberg, K. E. (1970). International news: Intra and extra media data.
Acta Sociologica, 13, 96-109.
Santons, T. D. (1996) The structure of dependence. In A. Inkeles & M.
Sasaki (Eds.), Comparing nations and cultures: Reading in a
cross-disciplinary perspective (p. 498-504). Englewood Cliffs, NJ: Prentice
Hall.
Shim, J. (2002) Deviance as the nature of international news: A concept
explication. A paper presented as the annual convention of the Association
for Education in Journalism and Mass Communication held in Miami Florida,
2002.
Shoemaker, P., Danielian, L. H., & Brendlinger, N. (1991) Deviant acts,
risky business and U.S. interests: The newsworthiness of world events.
Journalism and Mass Media Quarterly. 68, No. 4, 781- 795.
Shurnik, W. A. E. (1981). Foreign news coverage in six African newspapers:
The potency of national interests. Gazette, 28, 117-130.
Singer, Jane B. (2001). The metro wide web: Changes in newspapers'
gatekeeping role online. Journalism & Mass Communication Quarterly. 78 (1,
Spring), 65-80.
So, A. (1990). Social and development: Modernization, dependency, and world
system theories. Newsburry Park, CA: Sage.
Stephens, M. (1988). A history of news: From the drum to the satellite. New
York: Viking.
Wallerstein, I. (1974). The modern world system. New York: Academic Press.
Wanta, W., & Leggett, D. (1988). Hitting paydirt: Capacity theory and
sports announcer's use of clichιs. Journal of Communication, 38, 4, 82- 89.
Wu, H. D. (2000). Systemic determinants of international news coverage: A
comparison of 38 countries. Journal of Communication, 50 (2), 110-130.
<Figure 1> Shoemaker, Danielian, and Brendlinger's World Newsworthiness Model
Deviance Level of the Event
Social Significance Level of the Event
Low
Low
High
Proximity & Timeliness
NO
Proximity &
Timeliness
YES
Coverage of medium prominence
No coverage
Coverage of low prominence
High
Coverage of medium prominence
Coverage of high prominence
<Table 1> Significance points of 32 participating teams measured from the
FIFA points, William Hill betting points, and CNN Power Ranking
FIFA Ranking (points)
William Hill betting points
William Hill reversed rank order
CNN Power Ranking
CNN reversed rank order
Franc
802
4:1
32
2
31
Argentina
784
4:1
32
1
32
Brazil
784
6.5:1
28
4
29
Italy
717
5:1
30
3
30
Portugal
726
12:5
26
6
27
Spain
713
6:1
29
7
26
England
694
9:1
27
5
28
Germany
695
12:1
26
8
25
Cameroon
672
40:1
24
10
23
Paraguay
671
80:1
21
11
22
Croatia
655
80:1
21
9
24
Sweden
665
80:1
21
12
21
Ireland
674
80:1
21
16
17
Russia
644
66:1
23
13
20
Turkey
654
80:1
21
19
14
Denmark
657
125:1
11
13
20
Mexico
716
150:1
10
22
11
Belgium
653
100:1
15
15
18
Nigeria
644
100:1
15
17
16
Poland
615
66:1
23
17
16
Uruguay
652
100:1
15
20
13
Japan
634
80:1
21
21
12
United States
690
300:1
7
25
8
Slovenia
649
300:1
7
23
10
Ecuador
624
100:1
15
24
9
Costa Rica
643
300:1
7
28
5
Korea
603
150:1
10
27
6
Senegal
599
150:1
10
26
7
Tunisia
635
500:1
4
30
3
South Africa
623
500:1
4
29
4
Saudi Arabia
627
750:1
2
31
2
China
566
750:1
2
32
1
<Table 2 > Significance, Deviance, and Mixed Points for 48 matches
Significance
Deviance
Mixed
FIFA points
William Hill
CNN
Power
FIFA points
William Hill
CNN
Power
FIFA points
William Hill
CNN
Power
France : Senegal
1401
42
38
48
48
48
1604
63
63
Ireland : Cameroon
1346
45
40
24
23
28
1344
42
34
Uruguay : Denmark
1309
26
33
22
37
17
1304
30
26
Germany : Saudi
1322
28
27
10
5
3
1254
6
4
England : Sweden
1359
48
49
29
28
29
1330
42
42
Paraguay : S. Africa
1294
25
26
30
33
35
1246
8
8
Argentina : Nigeria
1428
47
48
16
11
9
1388
30
32
Spain : Slovenia
1362
36
36
11
5
9
1298
14
20
Croatia : Mexico
1371
31
35
12
43
44
1310
42
48
Brazil : Turkey
1428
49
43
3
16
11
1298
42
28
Italy : Ecuador
1341
45
39
5
13
6
1248
30
18
China : Costa
1209
9
6
9
18
20
1132
4
2
Japan : Belgium
1287
36
30
27
28
28
1268
30
24
Korea : Poland
1218
33
22
39
44
42
1230
46
32
Russia : Tunisia
1279
27
27
20
9
7
1270
8
10
USA : Portugal
1416
33
33
42
46
46
1452
52
52
Germany : Ireland
1369
47
42
28
25
30
1348
42
34
France : Uruguay
1454
47
44
36
33
35
1304
30
26
Cameroon : Saudi
1299
26
25
14
5
6
1254
4
4
Denmark : Senegal
1256
21
27
33
21
33
1198
20
14
Sweden : Nigeria
1309
36
37
19
17
18
1288
30
32
Spain : Paraguay
1384
50
48
15
15
20
1342
42
44
Argentina : England
1478
59
60
45
39
38
1568
64
64
S. Africa : Slovenia
1222
11
14
40
36
40
1248
14
20
Italy : Croatia
1372
51
54
43
42
40
1434
60
60
Brazil : China
1350
30
30
1
1
1
1132
4
2
Mexico : Ecuador
1340
25
20
6
38
22
1248
27
18
Costa : Turkey
1297
28
19
25
28
24
1287
22
17
Japan : Russia
1278
44
32
38
35
41
1288
46
40
Korea : USA
1293
17
14
34
23
24
1206
14
12
Tunisia : Belgium
1288
19
21
26
30
25
1270
8
16
Portugal : Poland
1341
49
43
4
19
14
1230
46
32
Denmark : France
1459
43
51
47
48
43
1604
64
62
Senegal : Uruguay
1251
25
20
32
25
28
1198
20
14
Cameroon : Germany
1367
50
48
17
20
22
1344
48
46
Saudi : Ireland
1301
23
19
13
9
11
1254
4
4
Sweden : Argentina
1449
53
53
35
30
31
1330
42
42
England : Nigeria
1338
42
44
31
31
32
1288
30
32
S. Africa : Spain
1336
33
30
7
2
4
1246
8
8
Slovenia : Paraguay
1320
28
32
18
14
13
1298
14
20
Brazil : Costa
1427
35
34
2
6
2
1286
14
10
Turkey : China
1220
22
15
8
9
12
1132
3
2
Italy : Mexico
1433
40
41
23
34
36
1432
20
22
Ecuador : Croatia
1279
36
33
41
40
45
1310
42
48
Japan : Tunisia
1269
25
15
37
11
15
1270
8
6
Belgium : Russia
1297
38
38
21
41
37
1288
46
40
Korea : Portugal
1329
36
33
46
45
47
1352
52
54
Poland : USA
1305
30
24
44
12
16
1380
14
16
<Table 3 > Number of articles and matches covered by the media
New York Times
Washington Post
Online version of New York Times
Online version of ESPN
Chosun-Ilbo
KBS
Online version of Chosun-Ilbo
Online version of KBS
News articles
35
47
44
61
93
59
97
49
Matches covered
44
58
65
67
112
69
124
53
<Table 4> "Importance of Coverage" Scores assigned to 48 matches from the
media coverage.
US media
US online version
Korea media
Korea online version
NYT
WP
NYT.com
ESPN.com
Chosun-Ilbo
KBS
Chosun.com
KBS.co.kr
France : Senegal
21
26
8
9
66
54
7
17
Ireland : Cameroon
2
6
3
2
12
0
3
0
Uruguay : Denmark
5
2
4
3
19
0
6
0
Germany : Saudi
6
4
5
5
31
0
5
12
England : Sweden
2
7
3
4
24
30
12
7
Paraguay : S. Africa
3
5
2
3
31
2
2
2
Argentina : Nigeria
4
6
5
5
21
15
9
0
Spain : Slovenia
2
7
2
4
34
8
3
5
Croatia : Mexico
0
4
4
3
11
0
3
0
Brazil : Turkey
7
6
5
5
17
21
9
7
Italy : Ecuador
6
2
4
2
18
0
9
0
China : Costa
6
5
5
2
19
12
7
0
Japan : Belgium
6
7
0
2
20
18
7
0
Korea : Poland
7
6
4
2
0
0
43
20
Russia : Tunisia
2
4
2
2
0
0
1
0
USA : Portugal
24
24
16
15
7
21
5
7
Germany : Ireland
5
6
3
2
22
0
8
2
France : Uruguay
6
5
7
11
31
8
9
8
Cameroon : Saudi
0
6
4
5
10
2
0
0
Denmark : Senegal
0
2
4
4
15
5
4
0
Sweden : Nigeria
2
6
4
3
12
0
3
0
Spain : Paraguay
4
4
4
2
11
15
4
0
Argentina: England
6
16
11
9
39
32
7
7
S. Africa : Slovenia
2
4
2
4
6
11
4
9
Italy : Croatia
4
11
3
5
13
14
6
0
Brazil : China
2
6
2
4
21
6
6
4
Mexico : Ecuador
0
6
4
2
11
9
2
0
Costa : Turkey
6
6
3
2
7
7
3
0
Japan : Russia
6
5
8
2
12
6
8
9
Korea : USA
22
5
26
13
0
0
40
15
Tunisia : Belgium
0
2
5
0
6
0
2
0
Portugal : Poland
5
4
5
0
9
0
5
0
Denmark : France
14
18
7
7
36
42
9
5
Senegal : Uruguay
0
2
4
6
11
9
1
2
Cameroon: Germany
5
4
4
5
13
13
5
0
Saudi : Ireland
6
6
3
4
0
5
4
3
Sweden : Argentina
6
12
5
5
21
18
7
0
England : Nigeria
6
0
0
5
14
16
5
0
S. Africa : Spain
0
0
4
3
16
0
1
5
Slovenia : Paraguay
4
0
3
4
11
0
5
9
Brazil : Costa
7
6
6
6
12
11
6
9
Turkey : China
4
4
4
5
11
6
3
5
Italy : Mexico
6
11
8
7
0
0
0
5
Ecuador : Croatia
0
5
0
7
11
0
0
0
Japan : Tunisia
4
5
0
2
6
6
4
0
Belgium : Russia
5
0
4
3
0
0
3
0
Korea : Portugal
22
12
5
6
0
0
36
14
Poland : USA
16
26
6
7
0
18
4
0
<Table 5> Regression analysis of newsworthiness factors and media coverage
about U.S. traditional media
Index
Newsworthiness factors
The New York Times
The Washington Post
Beta t
Beta t
FIFA points
Significance
.287 2.127*
.489 4.407**
Deviance
.382 2.831**
.481 4.341**
R
.235
.484
F
6.445**
19.695**
William Hill betting points
Significance
.202 1.426
.295 2.202*
Deviance
.348 2.456*
.377 2.811**
R
.189
.272
F
4.885*
7.852**
CNN Power ranking
Significance
.147 1.003
.288 2.120*
Deviance
.316 2.158*
.367 2.707**
R
.141
.263
F
3.454*
7.482**
*< .05 ** < .01
<Table 6> Regression analysis of newsworthiness factors and media coverage
about Korean traditional media
Index
Newsworthiness factors
Chousn-Ilbo
KBS
Beta t
Beta t
FIFA points
Significance
.395 2.858**
.419 3.509**
Deviance
.189 1.365
.440 3.686**
R
.206
.406
F
5.450**
14.354**
William Hill betting points
Significance
.287 1.951
.290 2.100*
Deviance
.156 1.062
.328 2.376*
R
.125
.230
F
3.006
6.273**
CNN Power ranking
Significance
.326 2.200*
.295 2.095*
Deviance
.115 .775
.302 2.150*
R
.139
.224
F
3.385*
6.074**
*< .05 ** < .01
<Table 7> Regression analysis of newsworthiness factors and media coverage
about U.S. online media
Index
Newsworthiness factors
NYT.com
ESPN.com
Beta t
Beta t
FIFA points
Significance
.508 3.867**
.515 4.354**
Deviance
.117 .892
.371 3.143**
R
.275
.414
F
7.979**
14.820**
William Hill betting points
Significance
.308 2.133*
.267 1.234
Deviance
.194 1.346
.189 1.816
R
.156
.126
F
3.875*
3.031
CNN Power ranking
Significance
.319 2.174*
.331 1.042
Deviance
.133 .907
.223 2.333*
R
.138
.191
F
3.349*
4.966*
*< .05 ** < .01
<Table 8> Regression analysis of newsworthiness factors and media coverage
about Korean online media
Index
Newsworthiness factors
Chosun.com
KBS.com
Beta t
Beta t
FIFA points
Significance
.423 3.029**
.266 1.799
Deviance
.076 .545
.124 .841
R
.191
.092
F
4.946*
2.140
William Hill betting points
Significance
.480 3.485**
.053 .336
Deviance
.027 .193
.076 .488
R
.236
.010
F
6.490**
.217
CNN Power ranking
Significance
.479 3.389**
.035 .223
Deviance
-.053 .377
.107 .673
R
.219
.015
F
5.877**
.311
*< .05 ** < .01
|
