|UFDC Home||myUFDC Home | Help|
This item has the following downloads:
THE TRANSPARENT GATE: ONLINE AND PRINT EDITIONS AT TWO CENTRAL
MATTHEW D. BLAKE
A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF ARTS IN MASS COMMUNICATION
UNIVERSITY OF FLORIDA
I would like to thank several individuals who provided support and guidance
throughout the writing of this study. First, I would like to thank Mindy McAdams, who
served as the chairperson of my thesis committee and offered her expertise concerning
online journalism. I would also like to thank committee members Dave Carlson and Dr.
John Sutherland for their valuable input.
I am also indebted to my parents, Philip and Katherine Blake, and brothers, Edward
and Daniel Blake, who provided loving support during the creation of this study. My
lifelong friend, Timothy Slack, provided much needed guidance throughout the graduate
TABLE OF CONTENTS
ACKNOWLEDGMENTS ........................................................................ ii
LIST O F TA BLES ........................................................ .................... ........v
LIST OF FIGURES .............................................................................. vi
A B S T R A C T .................................................................... .................. ........................ v ii
1 IN T R O D U C T IO N ......... ................. ............................................ .. ..................
O v erv iew ............................................................... ................................. ................... 1
Gatekeeping and the Internet Newspaper....................................................................3
R research P purpose .................. .......................................................... ............... ......
T h esis O utlin e .............. ................................................................ .............. .... .....
2 LITERATURE REVIEW................................................................7
O v erview ............................................................ 7
Gatekeeping in Mass Communication ................................. .... ..................7
G atekeeping and Social Reality ................................................... .................... 9
Electronic Newspaper Gatekeeping ...................................................................10
The Electronic N ewsroom ............................ ...... ........ .................... 14
Justification for Study ........................................ ................................ ................... .... 17
R research Q questions .................. .................. ................. .................... 19
3 M E T H O D O L O G Y ............. ........................................................ .................... 2 1
O v erv iew ...........................................................................2 1
Content Analysis Description......................... ...... .... ....................21
Study Sam ple ..................... ................... ........................... .................... 22
Study Sam ple: Print Edition .......................................................... .................... 23
Study Sam ple: Online Edition............................. .... ..... .................... 24
Sam pling Procedure ................................................................... .................... 26
Data Analysis ............................................... .................. 30
4 R E SU L T S ......... ......... .. ... ..... ... ................................ ....................32
L ocalization of E editions ..................................... ........................ .........................33
Reliance on N on-Staff Content ..............................................................................35
Use of Photographs and Artwork.........................................................................38
H deadline R relationship ................................................................ .................... 40
5 DISCUSSION AND CONCLUSION..............................................42
D discussion of R results .................................................................. ................... 42
Im plications of Study .................................................................. ................... 45
L im stations of Study .................................................................. .................... 47
Suggestions for Further Research .............................................. ....................48
C conclusion ..................... ..... ............................. ...... ... ... .........49
APPENDIX SAMPLE CODING SHEET............................................... ....................51
LIST O F REFEREN CES ......... ................... ......... ..........................................55
BIO G RA PH ICA L SK ETCH ................................................................ ...................59
LIST OF TABLES
4.1 Print and online story count by market .......................... ...... ....................33
4.2 Geographic location of story by market and edition............................................34
4.3 Article source by market and edition ................................... .................... 37
4.4 Graphic source by edition and market ........................... ................................38
4.5 Type of graphic by newspaper edition and market ..............................................40
4.6 Online Edition Headlines .................. ..................................... 40
LIST OF FIGURES
2.1 White's version of gatekeeping (based on Shoemaker, 1991, p. 10)..............9.....9
2.2 Inter-edition newspaper gatekeeping .......................................................19
Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Arts in Mass Communication
THE TRANSPARENT GATE: ONLINE AND PRINT EDITIONS AT TWO CENTRAL
Matthew D. Blake
Chair: Mindy McAdams
Major Department: Journalism and Communications
This thesis examines the relationship between the content in the print and Internet
editions of two central Florida newspapers, the Ocala Star-Banner and the Orlando
Sentinel. These two newspapers represent separate classes of circulation size and
corporate ownership while sharing circulation areas and local and state news stories.
This study is aimed at contributing to the base of knowledge concerning the role of the
online newspaper and its relationship to its print counterpart.
The researcher sought to determine the similarity in content in the two editions by
examining several variables. These included the localization of written content, source of
written and graphical content, type of graphical content made available and the article
headline text. These variables were examined using a quantitative content analysis.
This study found dissimilarities between the two editions. The Internet edition
offered a greater scope of state and local content, while the print edition provided more
national and global content. This study also found that articles in the print editions were
more likely to include a photograph than articles in the online editions and headlines were
nearly always replicated verbatim from the print to the online edition.
The study concluded that the focus of the online edition at the two newspapers is
more localized than the print edition and its articles seldom offer supplementary content.
If one begins with the reinterpretation of Plato's simile of the cave contained in
Public Opinion (Lippmann, 1922), the conveyance of information from the mass media
to its audience has been studied for nearly a century. Since Lippmann's idea of the
media's creation of a "picture in our heads," many theories have been initiated and
multiple innovations have assisted in the study of mass communication. Among
communication innovations, the Internet encompasses a unique audience and vast amount
of information. While television and print communications offer information from an
organization to the public, the Internet provides interactive communication to a global
audience. Like communication devices before it, the Internet has inspired scholarly
literature that searches to understand its use, its messages, and its effects on society.
Among journalists, the Internet presents a potential quandary and an opportunity.
While reaching a wider audience and providing advanced technology for transmission of
information, the newspaper Web site often channels information without demand for the
financial compensation that makes it a viable business and its information a valuable
commodity. Newspaper editors are faced with an industry-wide question of how to
present information on the Internet without compromising the traditional pay-for-access
format (Bolt, 2002; Geyskens, Gielens, & Dekimpe, 2002; Outing, 2002a, 2002b; Hall,
2001). This question has been answered in various manners. Some newspaper Internet
editions require payment for access to content, most notably The Wall Street Journal,
where in 1999, 50,000 users paid $59 annually for online access (Hall, 2001). Beyond
requiring payment for content, other newspapers offer different levels of access to
information: Some require registration for Internet content; some continue to offer online
content for free while charging for archived content; some offer entire current-day and
archived Internet content free while requiring payment for only the print edition. This
inconsistency tells of an evolving and developing model of media practice.
The costs of acquiring information online are often borne in the purchasing of a
computer and an Internet connection (Hall, 2001), not in direct payment to the
information provider. This shift in consumer costs has resulted in the newspaper's
reevaluation of how to present its information, at what costs, and on what medium. The
print edition is traditionally profitable because of its advertising revenue, which allows a
newspaper's financial viability (Bagdikian, 2000). A newspaper's print audience is what
maintains the print advertising revenue. The Internet edition has proved to be less
profitable in many cases it actually runs at a deficit. However, more than 3,000 U. S.
daily newspapers have Internet editions that run information similar to what is found in
the print edition (Palser, 2002). Past studies have shown different graphics, stories,
interactive features in each edition (Dibean, 1999; Gubman & Greer, 1997; Li, 2002;
Singer, 2001a, 2001b).
Exposure to each edition offers different exposure to what is considered
newsworthy. The different information traveling through each model of transmission
illustrates the concept of media gatekeeping: Who gets what information.
Gatekeeping and the Internet Newspaper
Gatekeeping as constructed by Lewin (1947b) is composed of "channels" that
"proceed in definite steps" (p. 144). Lewin uses the distribution and consumption of food
to describe this process. Channels to obtain food include the purchase at the grocery
market or growing vegetables in a garden. After obtaining the food, the next step in the
gatekeeping process is another channel option, to store or consume the food. The
progression of channels can continue to the preparation and consumption of the food.
Once material passes through a channel its properties are altered to reflect its
modification. Lewin's example is the status of food after purchase: "Having invested a
substantial sum of money in the food, she [provider] will be especially insistent that the
food safely reach the table and be eaten" (p. 145). This is an example of the "gate,"
which is a section in the channel that dictates whether the good passes through the
channel based on its properties prior to entering the channel.
The gate or gate sections are governed by impartial rules or by "gate keepers"
(Lewin, 1947b). The gatekeeper in this example is the person who either buys or
produces the food. Impartial rules, according to Lewin, can include economic and social
Lewin only briefly mentions gatekeeping's application to the study of
communications, but the widespread acceptance of Internet editions by newspapers to
complement print editions provides a unique opportunity to study communications'
gatekeeping function between editions. Lewin's food consumption model can be
replicated by substituting the material (instead of food, written and graphical content), the
gatekeeper (instead of the purchaser or provider of food, the newspaper editor), and the
gate (instead of the purchase of food, the transfer of information to the Internet or
newspaper from the newsroom).
In this revision of Lewin's gatekeeping model, the news content -both written and
graphical -is present in the newsroom from multiple sources: First, the newspaper staff;
second, the wire service or syndicate; third, freelance writers; and fourth, the readers of
the newspaper in letters to the editor. This information is judged by the gatekeeper
(editor) and is either passed through the Internet channel -and transferred to the
newspaper's online edition -or not passed through the channel to be placed online.
Using this model, the researcher intends to examine the content that is present in the print
and online editions of two newspapers.
Applying the idea of media gatekeeping to the Internet newspaper is not a new
concept, however, current research is limited. Further evaluation is required, if for no
other reason than the constantly changing nature of Internet news content and
presentation (McMillan, 2000). This study intends to demonstrate the news content
differences, in amount and display, in the print and online editions of the daily
newspaper. Both the print and online newspaper serve as gatekeeping channels through
which news and information travel to its audience; these channels stream different
amounts and types of information to each respective audience (Singer, 2001b; Gubman &
Greer, 1997; Schafer, 2002).
By examining the print and online editions of selected newspapers, the author
hopes to demonstrate the different "pictures in our heads" that each format presents. This
study will look at the news content of the print and online editions of two central Florida
newspapers and compare the results. Examining the content in the print edition and
comparing it to the online content allows comparison of what stories are redistributed
online and how articles appear online compared to the print appearance. By displaying
the variance in content, the author will demonstrate the contrasting perspective delivered
to the readers of a publication's online and its print edition.
The following chapter investigates the fundamentals of the gatekeeping process in
mass communication. The author reviews the seminal studies before narrowing the focus
to electronic and Internet gatekeeping. There are two established manners of
investigating newspapers' Internet presence: First, by looking at the content and format
of newspapers' Internet presence, and second, by researching the wired newsroom and
the process of channeling information from the print to the online edition. For the sake
of brevity, the latter will be reviewed but not analyzed; the focus of this study is the
articles presented in each edition (print or online). Chapter 2 will close with the study's
In Chapter 3, the researcher will present a quantitative content analysis model
used to collect data for this study. This study employed a composite week model that
investigated several variables. Chapter 3 will also discuss the sample used in the study,
the units of analysis, and the general methodology.
Chapter 4 will look at the findings of the quantitative content analysis of the print
and online editions of the newspapers. The findings will establish the dissimilarity
between the two editions of the selected newspapers in several categories. Differences in
content can be found in the geographic emphasis of the stories presented in each edition,
the source of the written and graphical content in each edition, the type of graphical
content used online and in print, and the phrasing of the headlines among identical
Chapter 5 will discuss the study's findings and discuss the implications and
weaknesses of the conducted research. Chapter 5 will also conclude the study with a
general conclusion of this study's significance.
The adoption of new technologies by established media demands reevaluation of
the production and presentation of information to its audience. When a newspaper
establishes an Internet presence, the effects are significant in personnel, financial
distribution, and most notably, the diversity of channels on which to transmit the
information that allows the organization continued viability. Media gatekeeping theory is
effective when evaluating the dissimilarity between content produced by traditional,
established print media and Interet-based media. This chapter begins with a review of
relevant academic research and literature, followed by research questions.
Gatekeeping in Mass Communication
As noted in chapter 1, the concept of a "gate keeper" was introduced by social
psychologist Kurt Lewin, who in 1947 coined the term while studying food consumption
habits. The gatekeeper in Lewin's study was the person who decided what food appeared
on a family's dinner table. In the process leading to the appearance of food on the dinner
table, the food passed through several "channels" or gate sections. Lewin briefly related
this process to the news selection process, which he noted must be governed by a
gatekeeper or "impartial" rules.
Lewin's seminal study was followed by White's introduction of Mr. Gates (1950),
the wire editor who functioned as the at a medium-sized Midwestern newspaper. Mr.
Gates served as the gate for the news that would appear on the front and "jump" pages of
the newspaper, and his decision was final. For this reason, White considered Mr. Gate's
decision-making to be the most important of the various "gates" of news selection that
each story passed through. White's study has since met criticism, however, for limiting
its analysis to a single wire editor (Bass, 1969), and for not addressing the personal
factors that influenced Mr. Gate's decision-making processes (Gieber, 1960).
Since Lewin's and White's initial studies, researchers have investigated other
channels in the process of media gatekeeping. Gieber (1960) sought to analyze the
gatekeeping function of newspapers by reviewing the "judgments and perceptions of
some of the persons involved in the transmission of news to the community" (p. 199). He
concluded that the reporter must judge "whether the information is sufficiently
important" (p. 200) to the audience while considering its newsworthiness. McQuail and
Windahl (1981) expanded on this approach by creating a hypothetical model that
demonstrated the transference of messages from the news source (N) to the media
gatekeeper, who selects messages deemed suitable for the audience (M).
Beyond the organization that delivers the information, the process of gatekeeping
can be narrowed to the individual consumer of information. Shoemaker (1991) defined
this perspective as a process "by which billions of messages that are available in the
world get cut down and transformed into the hundreds of messages that reach a given
person on a given day." This interpretation holds certain validity among contemporary
media consumers, who are offered far more choices of communication channels than
those of previous decades.
NEWS ITEM 2 "
SOURCE OF NEWS ITEM 3 AUDIENCE
NEWS ITEM IT
Figure 2 1 White's version of gatekeepmg (based on Shoemaker, 1991, p 10)
Gatekeeping and Social Reality
While contemporary media users have more sources of news and information than
in the past, one effect of receiving that information remains unchanged the construction
of social reality from the information provided by news sources When information is
judged newsworthy by editors and passed through the gate to the public, it influences the
public perception of what is considered important (Shoemaker, 1991) This canbe
achieved in mass media beyondjournalism Advertising can influence consumer
behavior by portraying a product as necessary or beneficial, government announcements
can inform the public of an array of information, from weather alerts to crime trends,
editonals can influence the public by making judgment on an issue with supporting facts
But news is 'the language in which most other claims to attention circulate" (Schudson,
2003), because it can strengthen public knowledge through unbiased reporting of facts
and circumstances without the appearance of conflict of interest between the source of
the information and the facts presented.
This distinction between journalism and other mass media gives news a unique
position in a democratic culture. Journalists are expected to perform watchdog duties on
public and private institutions that would otherwise remain unchecked. Because
journalism offers theoretically unbiased information, it gains greater respect among the
public. Schudson (2003) calls this "moral amplification": "A news story is an
announcement of a special kind. ... It announces to audiences that a topic deserves
public attention." The responsibility of the journalistic gatekeeper is to keep the public
The process of journalistic gatekeeping goes beyond than the model of general
consumption presented by Lewin (1947). Shoemaker (1991) provides an example of
gatekeeping on the individual:
The process of gatekeeping is the process of creating social reality: If an event is
rejected by the media I use, it probably will not become part of the social reality I
perceive. If the event is accepted and displayed prominently, then it may not only
become part of my version of social reality but it may strongly influence my view
of the world. (p. 27)
Because the public perception of issues is framed generally by news organizations and
specifically by gatekeepers, the news culture of the Internet that offers immediate news
and information, presents new responsibilities for journalists and the public.
Electronic Newspaper Gatekeeping
The introduction of recent computer technologies into the newsroom was first
examined by Garrison (1980), who studied the adoption of the word processor and its
effect on the gatekeeping process. Garrison found that despite new technologies, slot
editors relied on printed hard copy (paper) for the delivery to copy editors, who edited
hard copy content on word processors to create a soft copy (electronic) before sending the
story to production. The study concluded that while the technology's effect on the
newspaper's gatekeeping function was minimal, the computer eased editing while
facilitating the copy flow from reporter to production.
The use of photographs transmitted via the Associated Press wire was studied in a
gatekeeping study by Wanta and Roark (1992/1993). The researchers found that
photographic gatekeeping decisions depended largely upon the newspapers' tradition,
national trends, newspaper market size, and which news events provided a wide variety
The introduction of the Internet has provided a new perspective on the
gatekeeping process. It would seem the process of gatekeeping is rendered obsolete with
the Internet, a medium that allows consumers to organize personal interests without the
traditional gatekeepers to channel information. But the few studies that have explored
online gatekeeping suggest the role of the gatekeeper is not disappearing but evolving.
This revision of the gatekeeping process has placed nontraditional media representatives
as gatekeepers, including discussion list owners (Allen, 1994); online newspaper editors
(Singer, 2001); and Webmasters (Beard & Olsen, 1999).
One manifestation of the gatekeeper is the Webmaster, who Beard and Olsen
(1999) argue fulfills characteristics of the mass media gatekeeper. The Webmaster, like
the newspaper editor, makes decisions regarding the selection and presentation of
messages. Like other traditional media gatekeepers, the researchers found that the
Webmaster's skills and experiences influence content and selection of information when
performing the gatekeeping role (Beard & Olsen, 1999). Internet editors at smaller and
medium-sized newspapers are often expected to perform multiple duties, while larger
newspapers allow one employee to assume the role of Webmaster (Singer, Tharp and
Among journalists, the online gatekeeper plays a unique role. Unlike reporters or
traditional editors, online staffers are often understood to be "content providers" who
understand the technology necessary for putting the information on the Web (Lasica,
1997). The content that is replicated from the print to the online edition is often referred
to as "shovelware," or "content that was created for the print product and has simply been
shoveled on the Web" (Singer, 2001a). Many critics (Katz, 1994; Regan, 1995; Lasica,
1996) begrudge shovelware as lacking original content in a fresh medium and not fully
utilizing the technology available to online. This criticism has been investigated
repeatedly -to mixed results -over the past decade.
Gubman and Greer (1997) found inconsistencies among sites after analyzing 83
online newspapers of all circulation categories for content, structure and interactivity.
The researchers found that electronic newspapers offered content that was mainly
reproduced exactly from the print edition. There was no localization of national stories
and few changes in the linear storytelling format adapted from the print edition. The
researchers also found few sites using multimedia or interactive tools to assist in
storytelling. Gubman and Greer concluded that much of the criticism of online
journalism was well founded.
Singer (2001a) looked at six Colorado newspapers' online editions and examined
the content for geography (local versus non-local), Web-only content, and artwork. The
study found very little original content on the online editions. Of 2,455 total stories
analyzed, only 158 were Interet-exclusive, and 2,173 ran only in the print edition,
suggesting that "shovelware" continued to be prevalent at these papers. In the Web sites
studied, more space was devoted to metro and local news, proportionately, than to the
content offered in the print edition. Another characteristic of the online edition, according
to Singer, was the lack of appealing graphics. Singer noted that "despite the Web's
multimedia capabilities, many online papers are less visually enticing than their print
counterparts, at least in terms of information-conveying graphics" (p. 76). Singer's
analysis seems to agree with Gubman and Greer's and that of other critics -too much
shovelware, little original content, few multimedia applications.
Singer (200 ib) examined the online coverage of the 2000 local and national
elections by surveying editors at the largest newspaper in each of the 50 states and all
American newspapers included in the Newspaper Association of America's largest
circulation category (250,000 and greater). Realizing the Internet offers means of
enhancing political discourse as a two-way, interactive medium with tools such as polls,
e-mail, discussion forums and information-retrieval tools, Singer sought to measure the
difference in Internet utilization between the 1996 and 2000 elections. In the four years
separating the elections, the study found more original content (78.9% of the analyzed
sites offered Web-only election coverage), increased traffic to Web sites, and extensive
interactive features including ballot guides, detailed candidate profiles, precinct finders,
and archived poll results. The online newspapers also offered frequent updates that
permitted competition with television as the medium of choice (Singer, 2001). These
capabilities drew a greater audience to Internet news sites during Election 2000, when
nearly 20% of Americans reported getting campaign news online, according to the Pew
Research Center for the People and the Press (Schafer, 2001). Such statistics suggest that
the Internet is becoming a viable political news source and gatekeeping is indeed moving
beyond simple shovelware.
Arant and Anderson (2000) performed perhaps the most extensive survey of
newspaper editors regarding online media ethics in the contemporary environment.
Despite not focusing on Internet newspaper gatekeeping, the survey offers insights into
the amount of original content relative to shovelware at newspapers of various sizes.
With over 200 responses, the researchers built upon Singer's foundations and
reestablished that Internet newspapers are moving beyond repurposed content. Among
the wide-ranging sample (from less than 15,000-circulation to more than 200,000-
circulation newspapers), Arant and Anderson found only 20% offering no unique content
on the Interet. Additionally, 31% offered Internet-only content and 53% presented
Interet-only sections and features. The survey also provided insight into what is added
when a story goes online: 60% of editors said they added hypertext links; 30% changed
artwork and photos, and 23% changed the structure of the story (Arant and Anderson,
Electronic gatekeeping has been examined from the inception of wired content
and the introduction of the word processor to the current circumstances of the global
audience and interactive mass media. Beyond specific gatekeeping studies, the
researcher must examine the electronic newsroom, where important gatekeeping and
editorial decisions are made.
The Electronic Newsroom
The Internet offers the media professional new means of both collecting and
transmitting information. While researchers such as Garrison have written extensively on
reporters' use of the Internet for reporting, the purpose of this study is not to examine
reporter's use of information; therefore, media professionals' use of the Internet for
research will be ignored here. The focus of this section is the environment at Internet
newsrooms -how many employees, what tasks are employed, the general relationship
between the online and print newsroom and the disparities between large and small
newspapers. These variables allow for inconsistencies in online newspaper's content and
appearance, which is the focus of this study.
How to staff an online edition of a newspaper continues to be explored differently
at various institutions. For example, the New York Times houses its online staff in a
different building than the print staff, while The Wall Street Journal reconfigured its
newsroom to accommodate both staffs (Singer et al., 1999). Even among newspapers
with the same ownership, approaches differ. The Boston Globe and the New York Times
are both owned by the New York Times Company, yet take varied approaches at their
Internet identities. While the Times emphasizes similarity to its print edition, the Globe's
Web site, Boston.com, serves as a city guide as well as a media outlet.
There is perhaps no greater indication of technological and human resources than
the circulation of the newspaper and its operating budget. Garrison (1998) examined
technological discrepancies among newspapers of various circulation categories and
found that large newspapers (newspapers with a circulation greater than 50,000) hold
several advantages over smaller newspapers. Garrison (1998) found that large
Have more individuals involved in computer-assisted reporting than small
Are more likely to provide computer training than small newspapers.
Subscribe to "expensive" (p. 5) online databases (Lexis-Nexis, Database
Technologies' Autotrack Plus, Dow Jones) more often than small newspapers.
Devote a greater amount of financial support to online services.
While conceding the "common sense" (p. 6) findings of the study, Garrison did show
that large newspapers are able to grant more resources to Internet editions and computer-
assisted reporting than counterparts at smaller newspapers. During the time of the survey
(January through March 1997), more than 76% of larger newspapers, and fewer than 60%
of the smaller newspapers, had Internet editions (Garrison, 1998).
Singer et al (1999) investigated the responsibilities, attitudes and salaries of online
and print staffers and editors in her 1998 survey that looked at online and print
newsrooms. Like Garrison, Singer found that larger papers were likely to offer resources,
especially in terms of online edition employees, where smaller newspapers were more
likely to have crossover staff (employees performing tasks at two departments) from the
print to the online newsroom. The study also showed copy editors to be the most likely
to do double duty in the online and print newsrooms. Financially, print employees fared
better than online employees, and the sizes of online staffs were smaller than those of
print newsrooms (Singer et al, 1999). These disadvantages in the online newsroom
affected attitudes in both the print and online newsrooms, as evidenced in open-ended
responses to the survey. Several print managers described their job as harder because of
firm deadlines and space restrictions. Online managers said some of the problems they
face include finding qualified applicants for technological positions and the difficulty of
producing a profitable product. Online managers also complained about a lack of respect
from their print colleagues (Singer et al, 1999).
Justification for Study
The process of gatekeeping determines what information is provided to the public
from traditional journalistic organizations. Prior to the implementation of the online
edition at daily newspapers, only one channel from the print journalist to the public
existed to measure the gatekeeping process. With more than 1,200 daily newspapers
offering an Internet edition (Singer, 2001b), there are now two channels that can be
compared and analyzed.
The collective criticism of newspapers' Internet edition is the repurposing of
content, or "shoveling," from the print edition. The introduction of computer
technologies to the newsroom alters the editors' ability to dictate what information is
provided to the public to shape readers' social reality of what is newsworthy. If print
edition content is "shoveled" to the Internet edition, the process of discriminating
between content passed through this channel is eliminated. This holds important societal
consequences in a democracy that allows citizens to influence public policy. If the online
edition replicates print edition content, it follows that the information provided to the
audience is not unique to either newspaper edition. Past studies (Singer, 2001a; Li, 2002;
Gubman & Greer, 1997) have demonstrated that online editions are not precisely
replicated from the print edition and usually include a greater proportion of local stories
compared to national and global articles. These findings demonstrate the online
newspaper reader perceives a more localized reality of what is newsworthy compared to
the reader of the print edition.
The process of taking content from one edition and channeling it to another
format is an example of gatekeeping. The gatekeepers are the editors who decide what
print content is to be repurposed in the Internet edition. How the two editions differ
determines the extent of distinct information provided to the each audience. Three
manners in which the editions can differ are the written and graphical source of the
article, the geographic emphasis of the article and the supplementary interactive features
in the online article. This will be the focus of the current study.
This study will apply past models to the gatekeeping process linking the print and
Internet editions at two daily newspapers. By examining the print content and comparing
it to the Internet edition content, the researcher hopes to find what content is repurposed
in the Internet edition. Figure 2.2 illustrates the process that will be measured. The
original newspaper print edition content will be analyzed and the results will be used to
find what content was transferred to the Internet edition. This will be measured by
geographic emphasis and the source of written and graphical content
NEWS ITEM 1
NEWS ITEM 2
PRINT EDITION NEWS ITEM 3
CONTENT NEWS ITEM 4
Figure 2 2 Inter-edition newspaper gatekeeping
This study will demonstrate the different information provided to the audiences of
both the pnnt and Internet newspaper edition
The process of media gatekeepmg in both pnnt- and online-edition newspapers
has been investigated Singer (2001) offers a validated design for examining the
discrepancies between pnnt and online editions of newspapers This includes the
representation of the Newspaper Association of America's (NAA) circulation categones
and the use of a constructed week to measure vanables Using established methods of
comparing online and pnnt content, this study will ask the following
RQ1: Is there a relationship between geographical location of content and edition of the
newspaper (print versus online)? Is this relationship affected by market (metropolitan
RQ2: Is there a relationship between use of staff and non-staff content and edition of the
newspaper (print versus online?) Is this relationship affected by market (metropolitan
RQ3: Is there a relationship between graphical content and edition of the newspaper
(print versus online?) Is this relationship affected by market (metropolitan versus rural)?
RQ4: Is there a relationship between headline content and edition of the newspaper (print
versus online)? Is this relationship affected by market (metropolitan versus rural)?
To examine the mass media gatekeeping process, researchers rely upon different
methodologies, including the case study, the survey, the in-depth interview and the
content analysis. While each offers advantages, content analysis allows the researcher to
unobtrusively compare the exact content that passes through the "gates" or "channels" in
the process. This becomes especially relevant when comparing the content of a
newspaper's online and print edition because often the stories are repurposed
"shovelware" and lack overall creativity on the online edition (Katz, 1994; Regan, 1995;
Gubman & Greer, 1997; Singer, 2001b). Print and online editions have been compared
and researched during the past decade, but due to the Internet's continual adaptation of
new technologies and continuous changes in content (McMillan, 2000) past findings must
be reevaluated to measure current validity.
This chapter will begin with definitions and functions of content analysis,
followed by a detailed explanation of the researcher's selection criteria for newspapers,
time frame, and stories. After listing the coding categories, the conceptual and
operational definitions of key terms stated in the research questions will be provided.
Content Analysis Description
Berelson offered the most commonly used definition of content analysis: "Content
analysis is a research technique for the objective, systematic and quantitative description
of the manifest content of communication" (1952, p. 18). Krippendorff (1980) explains
content analysis as "a research technique for making replicable and valid inferences from
data to their context" (p. 21). Content analysis is by nature an unobtrusive technique that
prevents errors being included in the data due to the subject's knowledge of the study.
Potential subject factors that could invalidate a study include: Awareness, influences,
stereotypes, and experimenter-interviewer interaction (Krippendorff, 1980). Inferences
from content analysis involve the sender and receiver of the message, the communication
channel and the message itself.
To compare online and print content of newspapers, one must examine articles in
both formats. Two newspapers were selected and both formats were examined for six
variables. The units of analysis for this study were the individual articles in both the print
and the online format. Considering each format for a single story allowed for individual
analysis of a single story in two formats.
To ensure validity beyond a single week, the researcher employed the constructed
week format of sampling. To construct a composite week, the researcher randomly
selected a Sunday from the sample period, followed by random selections of a Monday,
Tuesday, Wednesday, Thursday, Friday and Saturday. This eliminates the possibility of
overrepresentation of certain editions of the week, such as the larger Sunday paper. Using
a composite week also allows greater generalization over time (Riffe, Aust, and Lacy,
1993). Both online and print editions of each newspaper were content analyzed during
each day of the composite week period. In this study, the composite week was
constructed during four weeks beginning Saturday, February 1, 2003, and ending
Saturday, March 1, 2003. During the seven selected dates, the researcher conducted all
coding for the print and Internet newspaper content.
This content analysis includes two central Florida newspapers, the Orlando Sentinel
and the Ocala Star-Banner. These newspapers offer contrasting ownership and
circulation. Both, however, share overlapping circulation areas, and subsequently
distribute similar local and state news stories and coverage.
The Orlando Sentinel has a circulation of 254,956 (Editor & Publisher Yearbook,
2002) and is owned by the Tribune Company, the second largest newspaper publisher in
the United States (behind Gannett). The Tribune Company owns 12 newspapers
including The Chicago Tribune, Los Angeles Times, South Florida Sun-Sentinel, and
Newsday (Hoover's Company Profile Database, 2003).
The Ocala Star-Banner has a circulation of 52,041 (Editor and Publisher, 2002) and
is owned by the New York Times Company. The New York Times Company includes
the New York Times newspaper, Boston Globe, the international Herald-Tribune, 18
smaller dailies (including the Star-Banner) and three weeklies (Hoover's Inc, 2003).
The contrast in circulation size is relevant because it has been demonstrated that
large newspapers offer greater resources for online productions than smaller newspapers
(Singer et al, 1999; Garrison, 1998). Ownership of the newspapers is relevant because
different corporate ownership indicates unique management philosophies in relation to
online and print publications.
Study Sample: Print Edition
The first articles to be selected and evaluated were those that appeared in the
current-day print edition of the two newspapers. For both newspapers, the front section
and local or state section were evaluated. In the Star-Banner, the local section is named
"Marion," the name of the county of publication; in the Sentinel, the local section is
named "Local & State." Other newspaper sections (sports, arts, entertainment, lifestyle,
business, travel, general feature sections, weather and advertising sections) were ignored.
Also withheld from examination were "In Brief' stories or updates because these
articles lacked image content and standard headlines. Determining what was an "In
Brief' story varied among the two newspapers. The Orlando Sentinel notated these
stories with a label reading "In Brief' and several one-paragraph stories. The Star-
Banner labeled these articles under a general subject heading (for example, "Nation" or
"World") and included multiple paragraphs. Other than the briefs and updates, each
current-day news story was coded in the two selected sections.
Study Sample: Online Edition
After coding stories from each newspaper's front and local sections, the researcher
examined the content of the same stories that appeared in the online edition. The Internet
content was examined on the same date as the print publications' release.
The method of finding the stories within the Internet edition included numerous
techniques. First, the researcher used the search function located on the home page of
each newspaper's online edition. Inside the search form, the researcher initially entered
the first four words of the article's print headline. For example, if the print headline read,
"Trauma centers warn lives could be at risk," the researcher entered "Trauma centers
warn lives" into the search form to examine results.
If the search results yielded stories included an identical headline or a subject
matter related to the print edition article, the researcher determined whether the story was
identical to the print edition by examining its first sentence and written content. If the
initial search failed to produce any results, the researcher then entered the entire headline
into the search form. The final step using the search function to find the story in the
online edition was to enter the article's reporter or author name, which resulted in a
display of the recent stories by the reporter or author. The search function was the most
commonly used method of finding online stories because it most often yielded articles
offered in the print edition.
If every search attempt failed to produce the story or a closely related article, the
researcher then manually searched the hypertext headlines appearing on the newspaper
home page and individual section home pages. This method differed for each newspaper
due to the each site's architecture.
The Orlando Sentinel offers five hypertext headlines on the top of the home page
(http://www.orlandosentinel.com/) and lists 20 additional hypertext headlines at the
bottom of the homepage under the label "More Headlines." Searching these headlines for
stories found in the print edition was the first step in manual search process. Next, the
researcher entered the "News" home page (http://www.orlandosentinel.com/news/) and
evaluated its hypertext headlines for stories found in the print edition. Besides evaluating
headlines on the "News" home page, the researcher investigated subsections contained in
the "News" home page navigation bar. This included examining the hypertext headlines
contained in the "Regional" section home page (http://www.orlandosentinel/news/
local/regional/), the "State" section home page (http://www.orlandosentinel/news
/local/regional/) and the Nation/World section homepage
The Ocala Star-Banner online contains a more basic site structure than the Sentinel
and thus required less intensive inquiry to find print stories that were not found with the
search function. The home page (http://www.starbanner.com/) offered three featured
headlines with the beginning of the articles and 10 hypertext headlines under the label
"AP Top Stories" which are updated periodically. These hypertext headlines were
searched first. The researcher then examined the "News" section
(http://www.starbanner.com/apps/pbcs.dll/section?Category=NEWS), which offered
hypertext headlines for local, state, national and global news.
Only after these steps failed to produce an online version of the print story would
the researcher concede that it was not included in the Internet edition of the newspaper.
The researcher did not count Interet-exclusive stories for two reasons. First, as
McMillian (2000) pointed out, the Internet is a constantly changing medium, one which
does not rely on static dates or times to produce stories or editions. There is also no
established time frame for which online stories remain posted in the online edition.
Without means of ensuring consistency among articles, the researcher could not promise
replicable results when finding Interet-exclusive articles.
The second reason is the process of gatekeeping, which evaluates the properties of
goods before and after passing through a channel or gate region. The channel in this
study separates the print edition story from the online story. The story appearing only in
the Internet edition never passed through this channel and therefore offers no comparable
counterpart appearing in the print edition. For these reasons, the researcher used a one-
way model of finding stories appearing in both editions.
During this composite week, both the print and online editions of the selected
newspapers will be analyzed. Every current-day news story will be coded; sports, arts,
entertainment, lifestyle, business, travel, general feature sections, weather and advertising
sections will be ignored. Artwork that accompanies selected news stories will be counted
as well. The following will be coded:
* The newspaper where the story appeared. This consisted of two values with two
* The edition where the story appeared. This consisted of two values with two
* The date of the article, expressed in month, day and year of publication. Also
recorded was the day of the week. Where xx=month, yy=day, zzzz=year,
* The type of images) that accompanied the story. The following values were
4. Composite Image
5. Photograph and Infographic
6. Photograph and Composite Image
8. Photograph, Composite Image and Infographic
9. Infographic and Composite Image
* The source of the images) that accompanied the story. The following values were
recorded -entries with multiple sources indicate multiple images and sources.
When an article contained two or more images, the researcher coded the source for
each and the results required multiple categories for image source.
2. Wire Service or Syndicate
4. Another Newspaper
5. Staff and Wire Service/Syndicate
6. Government (Police photos, NASA, etc.)
8. Staff and Contributor
9. Staff and Government
* The geographic emphasis of the story. This was divided into five categories: metro
(about something inside the county in which the newspaper is located); state (inside
Florida but outside the newspaper's home county); national (a national story,
outside of Florida); international (any story outside the United States). Some stories
could ultimately be coded in multiple categories; for example, the Elian Gonzalez
saga in 1999 was simultaneously an international, national and state story. In a case
of multiple locales, the story will be coded for each of the locations. The following
values were recorded
entries with multiple sources indicate multiple images and
5. Metro and State
6. Metro, State and National
7. State and National
8. National and Global
9. Metro and Global
10. Metro, National and Global
11. Metro, State, National and Global
* The source of the article's written content, or the writer's or reporter's affiliation.
The following values were recorded.
2. Wire Service/Syndicate
4. Another Newspaper
* The headline that accompanied the article. The print headline was first recorded
verbatim and then compared to the headline presented on the Internet edition (if
applicable). This required three values.
1. Print headline
2. Online headline same
3. Online headline different
* Six categories were created for online-only content which could not appear in print
format. These included video, audio, poll, interactive graphic, "E-mail story"
option and photo gallery. Each of these were recorded separately in the following
Utilizing the widely used statistical software program, Statistical Package for the
Social Sciences (SPSS), the researcher employed chi-square tests to measure levels of
Chi-square tests are used to describe relationships between variables by using a null
hypothesis (Babbie, 2001). The null hypothesis is defined by Babbie (2001) as "the
assumption that there is no relationship between two variables in the total population."
By accounting for the null hypothesis, the chi-square test reveals the likelihood that the
relationship between variables is real. The calculation of chi-square is based on the
comparison between expected and obtained frequencies and is used to "determine the
probability that the discovered discrepancy could have resulted from sampling error
alone" (Babbie, p. 487).
Tests of significance are commonly followed by an expressed level of
significance. The level of significance is the probability that the association between
variables could have been produced by sampling error, as opposed to what are considered
genuine scientific relationships. For example, a significance level of .05 denotes that an
association "as large as the observed one could not be expected to result from sampling
error more than 5 times out of 100" (Babbie, p. 492). The significance level used in this
study is less than .05 and permits the researcher to discard the null hypothesis at the 95%
This study sought to compare content of articles appearing in the online and print
edition of two newspapers in unique communities by asking the following research
questions: Is there a relationship between geographical location of content and edition of
the newspaper (print versus online)? Is this relationship affected by market (metropolitan
versus rural)? Is there a relationship between use of staff and non-staff content and
edition of the newspaper (print versus online?) Is this relationship affected by market
(metropolitan versus rural)? Is there a relationship between graphical content and edition
of the newspaper (print versus online?) Is this relationship affected by market
(metropolitan versus rural)? Is there a relationship between headline content and edition
of the newspaper (print versus online)? Is this relationship affected by market
(metropolitan versus rural)?
When examining these questions, the researcher examined the content of articles
in two sections, front and local, of newspapers varying in circulation volume and
corporate ownership. The researcher then searched for the articles in the online edition of
each newspaper. During the composite week of examination, the final story count for
both editions of both newspapers was 743. The print story count was 418 full-length
articles. Of these, 325 (77.8%) were offered in the online edition of the two newspapers.
The Orlando Sentinel's story count during the composite week of examination totaled
444 articles for both its print and online edition. The Ocala Star-Banner story count
during the same period totaled 299 articles for both editions.
The relationship between the individual newspapers and amount of content
present in the print edition and repurposed content for online use is important to examine.
If there is a significant difference in content between markets, this study would not be
able to compare the two newspapers with statistical validity. The findings indicate no
significant difference between the two newspapers studied. While a modest increase in
story count was evident in the larger market (Orlando), there is no evidence that this
varies because of market size. Because the difference between the two newspapers is
negligible in story counts in both editions, it allows for valid comparison.
Table 4.1: Print and online story count by market
Print % Online % Total %
Orlando 252 60.29 192 59.08 444 59.76
Ocala 166 39.71 133 40.92 299 40.24
Total 418 100.00 325 100.00 743 100.00
Chi square = .0111347030396494, df= 1, not significant
Localization of Editions
The first research question examined the emphasis on geographic region in the
print and Internet editions of both newspapers. As Singer (200 a) noted, the Internet is
perceived as a niche medium and the newspaper Web site offers the editor the
opportunity to provide unique local news on an international platform. The current study
sought to examine the relationship between location of story and redistribution of articles
from the print to the online edition of the Ocala Star-Banner and Orlando Sentinel.
When story location was compared with the edition (print and online) and the
individual newspaper, the relationship was statistically significant -- the geographic
location of the story does have a relationship to the story selection of online content. The
difference between the variables is strongest when looking at the Sentinel results, where
residual analysis shows a strong relationship between its print and online edition content
when looking at geographic location of stories. The greatest residual difference (-13.78)
exists in the value of the Sentinel's online global content, meaning that stories of a global
subject are less likely to be repurposed in the online edition than stories with a more
proximate geographic emphasis. This relationship continued when looking at the
Sentinel's online national content, which contained a residual value (-11.56) that
indicates fewer national stories were made available online than expected. Residual
analysis of the Sentinel's state (10.09) and metropolitan stories (5.77) demonstrated an
figure that exceeded expected values, which indicates more state and metropolitan stories
were made available than expected.
The geographic stratification of articles appearing in the Star-Banner does not
follow the pattern of providing more local content in the online edition. When looking at
residual analysis of global (6.98) and national articles (4.51), there exist a greater number
of articles than the expected value. Residual analysis of the state (-5.26) and
metropolitan articles (-0.06) show a value that falls below the expected number of stories.
The findings show that while the Sentinel's online content included more
metropolitan and state stories than expected, the Star-Banner reproduced fewer state and
metropolitan articles than expected. It appears that the NAA's largest circulation group
exhibit significant differences in geography of articles appearing in both the online and
print edition. At the smaller circulation newspaper a less-significant difference exists
between the two editions.
Table 4.2: Geographic location of story by market and edition
Orlando Orlando Ocala Print Ocala Total
Print Online Online
Global 41 18 35 29 123
% 33.33 14.63 28.46 23.58 100.00
National 77 44 51 43 215
% 35.81 20.47 23.72 20.00 100.00
State 61 61 45 30 197
% 30.96 30.96 22.84 15.23 100.00
Metro 38 36 22 21 117
% 32.48 30.77 18.80 17.95 100.00
Multiple 35 33 13 10 91
% 38.46 36.26 14.29 10.99 100.00
Total 252 192 166 133 743
% 33.92 25.84 22.34 17.90 100.00
Chi square = 28.2011616882942, df= 12, P <.05, significant
Reliance on Non-Staff Content
The second research question asked if there exists a relationship between use of
staff and non-staff content and edition of the newspaper (print versus online) and if this
relationship affected by market (metropolitan versus rural). This was measured using
two variables: First, the writer's affiliation (newspaper staff, wire service staff, or non-
staff contributor), and secondly, the source of photographs or graphic work
accompanying the article (staff, wire service, non-staff contributor, government source).
When both editions are counted, stories by writers not on the newspaper's staff
totaled 443 (59.6%), while stories by writers affiliated with the newspaper totaled 300
1 "Multiple" includes articles with more than one geographic classification, including Metro and State,
Metro, State and National, State and National, National and Global, Metro and Global, Metro, National and
Global, and Metro, State, National and Global
(40.4%). When looking at the source of stories within each edition, the contrast becomes
more apparent. As displayed in Table 4.3, the distribution of staff and non-staff material
across editions is substantial. Residuals for the Sentinel's print edition demonstrate a
greater-than-expected value of staff-written material (17.26) and a less-than-expected
value for non-staff material (-17.26). When looking at the Sentinel's online edition, the
residuals for the staff-written content (39.48) and non-staff-written content (-39.48) are
both considerable. This shows that while both editions of the Sentinel show a preference
for staff-written material, the online edition reproduces staff content much more
frequently than non-staff content.
The Star-Banner exhibits significant differences in the source of written content as
well. But where the Sentinel's online edition displayed more staff content, both the print
and online editions of the Star-Banner displayed greater proportion of non-staff material
compared to its staff content. Residual analysis of the print edition shows values of staff-
written material (-31.03) that do not meet expected value and non-staff material (31.03)
that exceeds expected value. Analysis of the online edition demonstrate a similar
relationship, with residuals below expected value for staff-written material (-25.70) but
above expected value for the non-staff material (25.70). This demonstrates that while
neither edition matches expected values of staff content, the Star-Banner online edition is
less dependent on non-staff material than the print edition.
Table 4.3: Article source by market and edition
Staff % Non- % Total %
Orlando 119 39.66 133 30.02 252 33.92
Orlando 117 39.00 75 16.93 192 25.84
Ocala 36 12.00 130 29.35 166 22.34
Ocala 28 9.33 105 23.70 133 17.90
Total 300 100.00 443 100.00 743 100.00
Chi square = 83.3379375665623, df= 3, P =.001, significant
Like the written content, the graphical content contained in both editions was more
likely to have been produced by a non-staff source than by the newspaper staff. Stories
with non-staff graphics constituted 21.66% of the entire sample, but within editions of
individual newspapers this distribution varied greatly. While the print edition relied
heavily on non-staff produced graphics at both newspapers, neither newspaper
reproduced a substantial number of non-staff graphics in the online edition. Residual
analysis of Table 4.4 supports this statement: Sentinel non-staff graphics in the print
edition (5.66) exceeded the expected value, while the non-staff graphics in the online
edition (-16.40) fell far below the expected value; Star-Banner non-staff graphics in the
print edition (24.20) far exceeding the expected value, while non-staff graphics in the
online edition (-13.46) fell below the expected value. This shows that the source of
graphics accompanying articles in the print edition differ from the source of graphics in
the online edition articles. It also appears that while both staff-produced- and non-staff-
produced graphics reduced in number when stories were transferred online, non-staff
2 "Non-staff' includes all sources not listed as newspaper staff, including Wire service and syndicate
articles, articles by contributors to the newspaper, and articles from other newspapers
produced graphics were less likely to reappear in the online edition than graphics
produced by the newspaper staff.
Table 4.4: Graphic source b edition and market
Staff Non-staff Combination4 None Total
Orlando 35 60 6 151 252
% 42.68 37.27 24.00 31.79 33.92
Orlando 24 25 19 124 192
% 29.27 15.53 76.00 26.10 25.84
Ocala Print 15 60 0 91 166
% 18.29 37.27 0.00 19.16 22.34
Ocala 8 16 0 109 133
% 9.76 9.94 0.00 22.95 17.90
Total 82 161 25 475 743
% 100.00 100.00 100.00 100.00 100.00
Chi square = 80.6193047840931, df= 9, p =.001, significant
Use of Photographs and Artwork
The use of artwork and images in the print and online editions was the focus of the
third research question of the current study. Already discussed was the frequency of the
sources of graphics in both editions; this section deals with the prevalence of the types of
graphics used in each edition. To measure this variable, the researcher classified
graphical content appearing in each article of both editions. The types of graphics
appearing in both editions were categorized as photographs, infographics, composite
images and cartoons. Since no cartoons appeared in the sample, this category will be
excluded from the following analysis.
3 "Non-staff' includes all graphic sources not listed as newspaper staff, including Wire service and
syndicate graphics, graphics by contributors to the newspaper, graphics from other newspapers and
graphics from government sources
4 "Combination" includes all articles that had graphics from multiple sources, including Staff and wire
service/syndicate, staff and government, and staff and contributor
When looking at the use of graphics it is important to first examine the distribution
between the two newspapers. Both newspapers' print editions contained more stories
without graphics than stories containing graphics and this reflects on the graphic use in
stories appearing in the online edition. Both newspapers offered visual content in almost
a third of all stories, and the majority of stories within both newspapers offered no in-
The relationship is altered when looking at stories in the individual online and print
formats. In the both newspapers the number of stories with photographic content
decreased significantly when moving from the print to the online format. The Sentinel's
residual analysis of the print edition (16.84) and online edition (-16.84) demonstrates that
stories containing photographs in the print edition did not usually contain photographs in
the online edition. The Star-Banner showed a similar, but stronger relationship.
Residual analysis shows the print edition (17.55) offering more photographic content than
the online edition (-17.55), while stories with no graphical content was far more prevalent
in the online edition (-23.27) than in the print edition (23.27). This shows that stories
with photographs in the print edition appear without photographs in the online edition.
Table 4.5: Type of graphic by newspa er edition and market
Print % Online % Total %
Orlando 90 21.53 40 12.31 130 17.50
Orlando 18 4.31 30 9.23 48 6.46
Orlando 144 34.45 122 37.54 266 35.80
Ocala 71 16.99 24 7.39 95 12.79
Ocala 8 1.91 0 0.00 8 1.08
Ocala 87 20.81 109 33.54 196 26.38
Total 418 100.00 325 100.00 743 100.00
Chi square = 46.8659442579277, df= 5, p <0.001, significant
The fourth research question sought to examine the similarity between headlines
appearing in the print edition and the online edition of the two selected newspapers. To
measure this variable, the researcher first recorded the article headline as it appeared in
the print edition and then compared its phrasing with the headline appearing in the online
edition. As Table 4.6 shows, among the stories reproduced in the online edition, the great
majority of headlines matched verbatim. Of 325 articles appearing in both editions, only
8 online articles used an original headline to accompany its written content. This
5 "Other" graphic types include all multi-format and non-photograph graphics including Cartoon,
infographic, composite image, photograph and infographic, photograph and composite image, photograph,
composite image and infographic, and composite image and infographic
6 "Other" graphic types include all multi-format and non-photograph graphics including Cartoon,
infographic, composite image, photograph and infographic, photograph and composite image, photograph,
composite image and infographic, and composite image and infographic
relationship, however, lacks statistical significance between editions and between
Table 4.6: Headline appearance by newspa er market
Print % Online % Online % Total %
Headline Headline Headline
Orlando 252 60.23 187 58.99 5 62.50 444 59.78
Ocala 166 39.77 130 41.01 3 37.50 299 41.22
Total 418 100.00 317 100.00 8 100.00 743 100.00
Chi square = 0.15131120724963, df
2, p < 1, not significant
DISCUSSION AND CONCLUSION
The purpose of this study was to investigate the variation of content among the
online and print editions of two central Florida newspapers. The research questions
asked the following: Is there a relationship between geographical location of content and
edition of the newspaper (print versus online)? Is this relationship affected by market
(metropolitan versus rural)? Is there a relationship between use of staff and non-staff
content and edition of the newspaper (print versus online?) Is this relationship affected
by market (metropolitan versus rural)? Is there a relationship between graphical content
and edition of the newspaper (print versus online?) Is this relationship affected by market
(metropolitan versus rural)? Is there a relationship between headline content and edition
of the newspaper (print versus online)? Is this relationship affected by market
(metropolitan versus rural)? A sample of the Orlando Sentinel and the Ocala Star-Banner
was collected during a composite week during March 2003. Both frequency and cross-
tabulation analyses were employed to answer the research questions. To determine
statistical significance, Chi-square tests were performed on the cross-tabulation tables.
Discussion of Results
The results demonstrated that among the two selected newspapers, a substantial
difference exists between the content presented in the online and print editions. During
the composite week period, each edition of the newspapers presented unique geographic
emphasis, different amounts of staff and non-staff produced stories and graphics, as well
as types of graphical content. Headlines were found to be almost identical in
When comparing the localization of online and print content, this study showed
that both newspapers focused on different geographical areas for online content. Content
originally appearing in the Sentinel print edition and redistributed to the online edition
was more likely to cover local and state issues rather than global or international issues or
subject matter. The Star-Banner's online content was concentrated on stories of a global
and national subject matter, rather than the more localized online content exhibited by the
Sentinel. This distinction is important because while the Sentinel followed the
expectations of previous research (Singer, 2001a), the Star-Banner demonstrated that
some newspapers do not necessarily place a greater local emphasis of online content.
Because of this distinction it appears that story geographic location is a selection
criterion for online distribution. For the Sentinel, the more localized the story, the more
likely it is to appear in the online edition; and conversely, for the Star-Banner, the less
localized, the greater the likelihood of an article appearing in the online edition. This
allows for important considerations of both audiences. Because the Sentinel audience is
exposed to a proportionately greater amount of local coverage and the print audience is
exposed to more global coverage, it would follow that each audience has a different
conception of what is "newsworthy" and perhaps a more localized social reality. The
Star-Banner's audience online would have a less-localized social reality compared to its
While the source of both graphical and written content in both newspapers leaned
toward non-staff content in both editions, differences existed in the redistribution rates of
both written and graphical content. The source of written content appears to have
affected selection for online distribution in the two newspapers, as stories reproduced
online were more likely to be written by a staff reporter than a non-staff writer. This is
true for stories with graphics, as well. While many articles were stripped of graphical
content when placed online, it was more likely that a staff-produced graphic would be
redistributed than non-staff graphical material.
There are three plausible explanations for this relationship. First, the editors
intentionally focus the online material on local issues, which happen to be covered by
staff reporters. Secondly, the required manpower of placement of articles on the Internet
may prohibit a complete replication of the print edition. Third, costs of distributing wire
service and syndicate content online may discourage redistribution of non-staff material.
The use of different types of artwork and graphical content also were dissimilar in
the two editions. While the majority of stories lacked accompanying graphics in both
formats, photographs were the most common type of graphic used and were more
commonly displayed in the print edition than the online edition. Despite lacking the strict
space limitations found in the print edition, the online edition did not display additional
graphics. This may be explained by the lack of resources devoted to the online staff
(Singer et al, 1999).
Headlines appearing in the online edition nearly always matched the print edition's
headline. Of the 325 stories appearing in both editions, only 8 stories (2.5%) contained
headlines that did not repeat the print headline verbatim. The level of homogeneity
between headlines of each edition suggests the use of content management system
applications to place stories in the online edition. The headline being an essential
element of the news article, the lack of variation among the two editions reinforces the
established criticism of online news being simply "shovelware."
Implications of Study
The process of media gatekeeping, which dictates what information is channeled
to the public, is essential in the functioning in a democratic culture. Because democracy
relies upon informed citizens to advance public policy, the media's gatekeeping is one
means of judging public exposure to issues and news events in society. Past research
(Singer, 200 la; Gubman and Greer, 1997) has demonstrated a unique function of online
newspapers when compared to the print edition. These studies have criticized the amount
of "shovelware" in the online edition and the amount of content repurposed from the print
edition. This criticism contends that since Internet editions lack much original content,
the newspapers' online editors fail to utilize the interactive and multimedia capabilities of
the Web and do not facilitate the gatekeeping function to inform the public of important
The current study is exploratory in nature and has several implications in regards
to the field of converged journalism and the ongoing transition from printed newspapers
to newspapers presented on the Internet. First, based on this study's results, it seems the
process of online "shovelware" is as prevalent as ever. Stories that were repurposed to
the Internet edition often varied little from the print edition other than the medium of
which the information was passed. This supports the findings of Singer (200 la) and
suggests that online editors are not fully utilizing the interactive and multimedia
capabilities offered by the medium. The headlines were mostly identical; the
photographs were mostly not transferred to the online edition; only major stories
generated interactive features and these were repurposed from other stories and other
These findings suggest an evolving notion of newspaper gatekeeping. Beyond
deciding what information is granted to the public, the Internet allows editors to choose
how to present newsworthy information. The online news story allows the capability to
include additional photographic, textual and multimedia content that is not viable in the
print edition, providing the audience with an expanded reality of what constitutes a news
This study is significant on two levels. First, it demonstrates that stories appearing
in both print and online editions show little difference in the written content and often
limits photographic content. Headlines and story content varied little between the online
and print formats. Photographs accompanying stories appeared more frequently in the
print format than the online format, despite the Internet's less restrictive space
limitations. The "gate" from the print to the online format is primarily a geographic filter
that allows much local and state content to appear while serving as a barrier to less-local
news being redistributed online. The "gate" also filters the source of content,
reproducing most staff content while limiting the redistribution of non-staff material. On
this basic level, the online gatekeeper is concerned with simple replication from print to
online format, while considering factors that determine a story's reproduction.
On another level, the online gatekeeper plays an important role in dictating, in the
case of significant news, how the story is told. The Shuttle Columbia story appearing in
the online Sentinel demonstrates this finding. The Columbia story had many local angles
from the Orlando area and generated 19 articles during the seven-day content analysis.
All 19 were reproduced in the online edition, representing 9.9% of all online content
replicated from the print edition of the Sentinel. Each online Columbia article was
accompanied with a photo gallery, video content, an interactive graphic and a "mail-to"
link, none of which were offered in the print edition. These features allowed a matchless
format, providing the audience and public with a truly multimedia understanding of the
event. In the case of the Columbia story, the repurposed content was textual and
photographic; however, multimedia and interactive content supplemented this material.
This demonstrates the gatekeeping role of the online journalists at the Sentinel
surpasses simple replication of the print edition. While the majority of articles online
lack interactive features, certain subjects offer opportunities to expand coverage beyond
what is capable at traditional media outlets. This manifestation of gatekeeping suggests
an additional gate present at online newspapers, a gate determining whether a story
deserves expanded interactive coverage or textual replication from the print edition.
Limitations of Study
The study has several limitations, the first of which is its small sample size. Most
studies examine more than two newspapers to generate findings of value and statistical
significance. This restriction threatens the external validity of the study and its larger
application. A study lacking external validity cannot be projected to other populations.
In this case, the results of the study are not applicable to other online and print
newspapers of the circulation range of the Star-Banner and Sentinel. The two
newspapers only apply to two -50,000 to 100,000 and more than 250,000 -of the four
NAA circulation categories.
Another limitation is the method used for data collection. Because the analysis
only included the front and local/state sections, it cannot be generalized to the
newspapers and their online websites in totality. Editorial, sports, entertainment and
other sections may demonstrate different results of the gatekeeping process. While this
study thoroughly examined the specified material, the scope could be increased both
within the selected newspapers and in the number of newspapers examined.
Suggestions for Further Research
While this study contained weaknesses in regard to external validity, it succeeded
in adding to the literature of online gatekeeping and offers opportunities for further
research. Ideally, a study of online newspaper content would include a sample that
incorporates all national regions, newspaper circulation sizes and major ownership
groups. While this would be difficult, a study examining a larger sample of any of these
categories would result in greater generalization among the specified population.
The study of ownership groups would provide findings to distinguish to levels of
interactive and multimedia content among different media companies. An examination
of newspaper circulation sizes could investigate the relationship between circulation size
and the amount and forms of content available online. A study looking at newspapers in
unique regions could examine the focus of online content in relation to community and
Beyond simply increasing the scope of the population sample, further research
could investigate the online-only content at daily newspapers. Because each media
organization has unique management and publication philosophies, online content varies
among each organization. This could be examined for a number of variables: multi-
media content, interactive content, community elements and the distribution of content
among different newspapers within the organization.
Further research could also investigate how decisions are made within the
newspaper management about what content is included online and what is withheld. This
could include a survey of editors' use of content management programs to transfer stories
to the online edition.
When considering the process of gatekeeping at journalism organizations, it is rare
to find an example of a study that examines the process occurring between two forms of
media. Past studies analyzed gatekeeping at the individual decision-making level
(Garrison, 1980; Gieber, 1960; Shoemaker, 1991) or at the communications routine level
(Martin, 1998; Shoemaker & Reese, 1991). This study analyzed the gatekeeping process
across different media with similar content ownership by following the model established
by Singer (2001a).
Gatekeeping provided an appropriate model to compare the content of the
newspapers' original model of information transmission (print) to the online model of
transmission. Using Lewin's (1947b) model of transference of goods between channels,
this study was conducted to compare the content of the online and print editions of two
central Florida newspapers. The purpose of this study was to explore the different
representations of news content provided to the consumers of online and print versions of
The results demonstrated a considerable dissimilarity between the online and print
editions of the two newspapers and the possible existence of a "gate" that dictates which
content is distributed online and which remains as print-only. While one newspaper
chose to focus online content on local and state issues, another focused on more global
and national issues. The newspapers also differed on the distribution of the sources of
graphical and written content. But at both newspapers, some elements changed little
from the print edition -the headlines were reproduced nearly always verbatim and little
original content existed in the online edition.
These findings offer important implications for the consumers of these news
sources. While examining only a fraction of newspapers and media in general, this study
does show that factors can determine what content is transferred across platforms. With
emerging technologies, news content becomes increasingly converged and content
becomes more easily redistributed between different media. This study provides only a
glimpse of current sharing of information between media.
SAMPLE CODING SHEET
News story identification #: #
1. Orlando Sentinel
2. Ocala Star-Banner
News story dated: Month
News Story First Sentence:
Type(s) of images) accompanying story: a.
4. Composite image
Image sourcess: a. b. c.
2. Syndicate or wire service (Associated Press, New York Times, Scripps Howard,
4. Another newspaper:
5. Parent company
Types of interactive features accompanying online story
(note all applicable):
3. "E-mail to" option
4. Photo gallery
6. Interactive graphics
Geography of story (note all applicable): a.
1. Outside of United States
2. Inside U.S.; outside Florida
Dateline (city, state):
Story writer name:
2. Wire service or syndicate (Associated Press, New York Times, Scripps Howard,
LIST OF REFERENCES
Allen, C. A. (1994, August). Internet hsts and gatekeeping. Paper presented at the
Association for Education in Journalism and Mass Communication Convention,
Arant, M. D., & Anderson, J. Q. (2000, August). Onlne media ethics: A survey of U.S.
daily newspaper editors. Paper presented at the Association for Education in
Journalism and Mass Communication Convention, Phoenix, AZ.
Babbie, E. (2001). The practice of social research (9t ed.). Belmont, CA: Wadsworth.
Bass, A. Z. (1969). Refining the "gatekeeper" concept: A UN radio case study.
Journalism Quarterly, 46, 69-72.
Bagdikian, B. H. (2000). The media monopoly. Boston, MA: Beacon Press.
Beard, F., & Olsen, R. L. (1999). Webmasters as mass media gatekeepers: A qualitative
exploratory study. Internet Research, 9 (3), 200-211.
Berelson, B. (1952). Content analysis in communication research. Glencoe, IL: Free
Bolt, J. A. (2002, August 16). Exec: Future of newspapers is online. AssociatedPress
Onlne. Retrieved September 12, 2002, from the LEXIS-NEXIS database.
Deuze, M. (2002, July 11). The Internet and its journalism, part I: A typology of online
journalism. Onlne Journalsm Review. Retrieved August, 26, 2002, from:
Dibean, W. (1999, March). How U.S. daily newspapers are using their Internet
counterparts. Paper presented at the University of Luton Creativity and
Consumption Conference, Luton, UK.
Editor and Publisher (2002). Editor and Publsher International Year Book (82nd ed.).
Garrison, B. (1980). The electronic gatekeeper: Editing on the copy desk
of a metropolitan newspaper. Newspaper Research Journal, 1(3), 7-17.
Garrison, B. (1998, August). Newspaper size as factor in use of computer-assisted
reporting. Paper presented at the Association for Education in Journalism and Mass
Communication Convention, Baltimore, MD.
Geyskens, I., Gielens, K., & Dekimpe, M. G. (2002, Spring). The market valuation of
Internet channel additions. Journal ofMarketing, 66 (2), 102-120.
Gieber, W. (1960). How the "gatekeepers" view local civil liberties news. Journalism
Quarterly, 37, 199-205.
Gubman, J., & Greer, J. (1997, August). An analysis ofonhne sites produced by U.S.
newspapers: Are the critics right2 Paper presented at the Association for Education
in Journalism and Mass Communication Convention, Chicago, IL.
Hall, J. (2001). Onlnejournalism: A critical primer. London: Pluto Press.
Hoover's Inc. (2003). Hoover's Company Profile Database American Public
Companies. Retrieved April 20, 2003, from LEXIS-NEXIS database.
Katz, J. (1994, September). Online or not, newspapers suck [Electronic version]. Wired, 2
Krippendorff, K. (1980). Content analysis: An introduction to its methodology. Beverly
Hills, CA., London: Sage Publications.
Lasica, J.D. (1996, November). A new concept in news. American Journalsm Review, 18
Lewin, K. (1947a). Frontiers in group dynamics: Concept, method and reality in science;
social equilibria and social change. Human Relations, 1, 5-40.
Lewin, K. (1947b). Frontiers in group dynamics: II. Channels of group life; social
planning and action research. Human Relations, 1, 143-153.
Li, X. (2002, Winter). Web page design affects news retrieval efficiency. Newspaper
Research Journal, 23 (1), 38-49.
Lippman, W. (1922). Public opinion. New York: macmillan.
Martin, S. (1994, Spring). External information databases in small circulation
newsrooms. Newspaper Research Journal, 15 (2), 154-160.
Martin, S. (1998). How news gets from paper to its online counterpart. Newspaper
Research Journal, 19 (2), 64-74.
McQuail, D., & Windahl, S. (1981). Communication modelsfor the study of mass
communications. New York: Longman.
McMillan, S. J. (2000, Spring). The microscope and the moving target: The challenge of
applying content analysis to the World Wide Web. Journalism and Mass
Communication Quarterly, 77, 80-98.
Mueller, J., & Kamerer, D. (1995). Reader preference for electronic newspapers.
Newspaper Research Journal, 16 (3), 2-13.
Nicholson, J. (1999, May 22). Miami Herald competes with Spanish offering. Editor &
Publsher, 132 (21), 151.
Outing, S. (2002a, January 9). Use Web to supplement your print edition. Editor &
Publsher, 132 (2).
Outing, S. (2002b, July 17). Knight Ridder Digital cedes some control. Editor &
Publsher, 132 (29).
Palser, B. (2002, November). We've only just begun. American Journalsm Review, 24
Regan, T. (1995, Winter). News alone is not enough. Nieman Reports, 49 (4), 78.
Riffe, D., Aust, C. & Lacy, S. (1993). The effectiveness of random, consecutive day and
constructed week sampling in newspaper content analysis. Journalsm and Mass
Communication Quarterly, 70 (1), 133-139.
Schafer, A. (2002, April). 2000 fizzled as the Internet election. Onlne Journalsm
Review. Retrieved August 27, 2002, from
Schudson, M. (2003). The sociology of news. New York: W. W. Norton & Company.
Shoemaker, P. J. (1991). Gatekeeping. Newbury Park, CA: Sage Publications, Inc.
Shoemaker, P. J., & Reese, S. D. (1991). Mediating the message: Theories of influences
on mass media content. New York: Longman.
Singer, J. B. (2001a). The metro wide web: Changes in newspapers' gatekeeping role
online. Journalsm and Mass Communications Quarterly, 78 (1), 65-80.
Singer, J. B. (2001b, September). Online newspapers go beyond shovelware. Paper
presented at the Association for Education in Journalism and Mass Communication
Convention, Washington, DC.
Singer, J. B., Tharp, M. P., & Haruta, A. (1999). Online staffers: Superstars or second-
class citizens? Newspaper Research Journal, 20 (3), 29-47.
Thomson Financial. (2003). Nelson'spublic company profiles. Retrieved April 20, 2003,
from LEXIS-NEXIS database.
United States Census Bureau. (n.d.a). 100 largest gaining counties. Retrieved September
3, 2002, from http://eire.census.gov/popest/data/counties/tables/CO-EST2001-
United States Census Bureau. (n.d.b). Florida Quickfacts. Retrieved September 5, 2002,
United States Census Bureau. (n.d.c). Estimated state demographic components.
Retrieved September 8, 2002, from
Wanta, W., & Roark, V. (1992/1993). Which wirephotos are used and which are rejected
at three newspapers. Newspaper Research Journal, 13/14 (4/1), 84-98.
White, D. M. (1950). The "gate keeper": A case study in the selection of news.
Journalism Quarterly, 59, 60-65.
Matthew D. Blake was born in Evanston, Illinois, in 1976. He spent his early
childhood in Madison, Wisconsin, before moving to Missoula, Montana, where he lived
for much of his upbringing. In 1998, Blake graduated from the University of Wisconsin
Madison, where he earned a Bachelor of Science and majored in political science.
Blake has instructed two Internet communications classes during his tenure at the
University of Florida, as well as creating Web sites for clients at the university and
beyond. His research interests include online communications and popular music history.
Blake begins the doctoral program at the University of Florida's College of Journalism
and Communications in Fall 2003.