Opening the Umbrella:
An Economic Analysis of Online Newspaper Geography



Hsiang Iris Chyi
Ph.D. Candidate
Department of Journalism
The University of Texas at Austin
Austin, TX 78712
Phone: (512) 445-0461
Fax: (512) 471-7979
E-mail: [log in to unmask]
[log in to unmask]

and

George Sylvie
Associate Professor
Department of Journalism
The University of Texas at Austin
Austin, TX 78712
Phone: (512) 471-1783
Fax: (512) 471-7979
E-mail: [log in to unmask]


Paper submitted to the Media Management and Economics Division,
1999 AEJMC Convention, New Orleans





Abstract

This study examines how the print newspaper's local nature and the Internet's
boundary-transcending capacity define an online newspaper's geographic market.
An original theory of online newspaper geography is developed, and an
illustrative model proposed, from which inquiries about the nature of the new
medium are made. An email survey of online newspapers was conducted to test the
hypotheses and to provide valuable information about online newspaper geography.



Having realized the potential of the Internet as an emerging medium, the
newspaper industry responded by launching electronic editions of their
newspapers on the Web. As of April 1998, more than 750 Northern American dailies
offered online services[1] and 98 of the top 100 newspapers provided online
products (Newspaper Association of America, 1999). Worldwide, as of February
1999, almost 3,400 newspapers provided online services, of which more than 2,000
are based in the United States[2] (Editor & Publisher Interactive, 1999). Over
the past few years, the emergence and continuing growth of a substantial online
newspaper industry deserves scholarly attention.
The Internet, as a news medium, possesses many unique characteristics that have
redefined how news is produced, distributed, and consumed; for example,
interactivity, unlimited space, hypertext, multimedia presentation, search
capability, timeliness, and global reach. While each attribute represents a
certain facet of this online media revolution, this study chose to focus on the
Internet's boundary-transcending capacity. Much of the mystique of the Internet
lies in its global nature and this may be where the Internet will make a
fundamental change to human communication. So the following analysis explores
the geographic aspect of the online medium.
An online newspaper is created by the convergence of two media: the newspaper
and the Internet. Technologically, the Internet enables online newspapers to
seek a worldwide market. Practically, most online newspapers are owned and
operated by their print counterparts, which also serve as online editions'
primary content providers (Tankard & Ban, 1998; Fitzgerald, 1997). Such a
relationship creates a scenario that makes the online newspaper market difficult
to define. While the market for the print newspaper almost always is defined by
geographic boundaries (Lacy & Simon, 1993), the Internet is an inherently global
medium. Therefore, between the most local and the most global, online newspapers
may define their geographic markets differently and blurring of market
boundaries may occur.
Therefore, this study examines how the print newspaper's local nature and the
Internet's boundary-transcending capacity define an online newspaper's
geographic market. An original theory of online newspaper geography is
developed, and an illustrative model proposed, from which inquiries about the
nature of the new medium are made. An email survey of online newspapers was
conducted to test the hypotheses and to provide information about online
newspaper geography.
Theories and Inquiries
Media function in a "dual-product market" (Picard, 1989) -- the "information
market" for media goods, and the "advertising market" for access to audiences.
Both are contained within a specific geographic area, the boundaries of which
often are determined by access to the commodity. But Internet access has few
pure geographic limits. Any information posted on the Web is accessible to a
practically global audience. Chyi and Sylvie (1998) indicated that the medium's
global capacity may prompt online newspapers to seek markets on a number of
geographic levels, serving multiple audiences and competing with multiple media
within multiple geographic areas. Considering the complexity and variations at
different levels of geographic markets, they distinguished an online newspaper's
local market from its long-distance market based on the availability of the
online newspaper's print counterpart. A local market is where the online edition
and the print edition are both available, while a long-distance market is where
only the online edition is available.
This local versus long-distance distinction makes a general statement about
online newspapers' submarkets without differentiating various types of online
newspapers. According to the model, every online newspaper has a local market
within the print edition's trading area and a long-distance market that covers
the rest of the world. To closely examine online newspaper geography, in
addition to the local/long-distance dichotomy, the following analysis further
differentiates online newspapers' geographic markets by introducing a
hierarchical model of online newspaper geography.
Previous studies have shown that content-sharing between a newspaper's print and
online products is prevalent -- the print edition often serves as the primary
content provider for the online edition and thus similar or identical
information is published in two formats (Fitzgerald, 1997; Tankard & Ban, 1998).
Given content-sharing, people who are interested in reading a newspaper's print
edition probably also are interested in reading that paper's online edition
because of their interest in the new and information provided. In the print
newspaper market, circulation penetration usually is strongest in the
newspaper's home community. This also may hold true for the online edition. For
instance, Houston residents are interested in reading the print edition of the
Houston Chronicle and also might be interested in reading its online edition.
On the other hand, the Internet's boundary-transcending capacity enables a
worldwide audience to access any Web sites and thus expands the online
newspaper's market boundary beyond the print edition's circulation area. But
news, especially local news[3] -- unlike CDs, computers, or Hollywood movies --
generally is not a universal product. Excluding those in the primary circulation
and coverage area of a print newspaper, who would be interested in local news
offered by that particular paper? For example, besides Austin residents, who
would be interested in knowing what's happening in Austin, Texas? To do so, the
long-distance online reader would have to have connections or to be related
somehow to Austin: Some may be within Texas, probably fewer in other large U.S.
cities, and even fewer in Europe, Asia, or Africa. But a national electronic
newspaper -- for example, the online New York Times -- may attract a global
audience. In other words, a local newspaper may become a regional newspaper
online, and a national newspaper may become international online. However, it
may not be as easy for a local newspaper to become international online, unless
there are existing or potential niche markets in which a local newspaper
specializes. (For example, the San Jose paper may find a niche at the
international level by its coverage of the high-tech industry in the Silicon
Valley).
This study tries to increase the sensitivity of Chyi & Sylvie's typology (1998)
in dealing with online geography. A review of media economics literature found
Rosse's umbrella model of newspaper competition (1975) a delicate analytical
tool for geographic market definitions.
The "umbrella competition" model (Rosse, 1975) depicted intercity competition
among (print) newspapers at four different layers in a metropolitan area --
metropolitan dailies, satellite city dailies, suburban dailies, and weeklies.
Despite the focus on intercity competition in the print newspaper market,
Rosse's model conveys a strong geographic emphasis by the hierarchical structure
of umbrellas of various sizes covering newspaper markets. Although the current
study does not intend to apply that model to the online newspaper market, it
uses the umbrella symbol to illustrate online newspapers' geographic markets
because it effectively captures the geographic aspect of a market definition.
The umbrella symbol serves here as a useful tool to illustrate a (print or
online) newspaper's market boundary and market size, as well as the particular
market's relative position in a larger geographic context.
To better describe online newspapers' geographic markets, a five-layer model is
created -- including community, metro, regional, national, and international
levels. Most print newspapers operate within the community, metro, or national
layer. The international layer is added because the Internet is a global medium,
and so is the regional layer because it is highly possible that between a
metropolitan market and the national market a regional audience exists. These
layers exist in a relative sense, meaning there is no clear-cut demarcation
between layers. In general, this new umbrella model illustrates print and online
newspapers' geographic markets. (Illustration 1).


















Illustration 1:         The Umbrella Model of Online Newspaper Markets
(Taking a metropolitan newspaper's online edition as an example)
 Online Newspapers in the Local Market
Because of content-sharing between a newspaper's print and online editions and
the local audience's interest in local information, it is logical if online
newspapers attempt to look for an audience within the print edition's market.
For example, the print edition of the Houston Chronicle is the dominant
newspaper in the Houston area. Its online product -- also featuring news and
information about Houston -- may find it easier to appeal to Houston residents
interested in information about Houston. In such cases, an online newspaper's
local (as opposed to long-distance) market boundary seems to be predetermined by
the print edition's market boundary. The formation of a local market definition
can be phrased using umbrella terminology: The print edition "transfers" its
original umbrella to the online edition. It is hypothesized that
H1a: An online newspaper's local market boundary is defined by the print
edition's market boundary. (The Umbrella-Transferring Hypothesis)

In addition to market boundaries, it also is reasonable to suggest that the size
of a print newspaper's market is positively related to the size of the online
edition's local market. A larger print market implies a larger group of people
interested in news and information provided by that newspaper. These people are
potential online audience for that same newspaper's online edition. Therefore,
it is hypothesized that
H1b: The larger the print edition's market is, the larger the online edition's
local market is. (The Umbrella-Correlation Hypothesis)

Online Newspapers in the Long-Distance Market
The umbrella-transferring and umbrella-correlation hypotheses specify the
relationship between the print edition's market and the online edition's local
market. As discussed above, online newspapers can go beyond that local boundary
to reach a long-distance market, where the print edition is not readily
available. Because of the medium's global nature, online newspapers can easily
reach audiences outside the print edition's circulation area and choose to seek
a market at any geographic level -- regional, national, or even global. However,
while technologically viable, the nature of the product often imposes
limitations on online newspapers' geographic market definition because news
generally is not a universal product. As discussed earlier, the Internet may
"upgrade" an umbrella from community to metro, from metro to regional, or from
national to international, but probably not from community to international.
Therefore, it is hypothesized that
H2: An online newspaper's long-distance market is one level above the print
edition's market. (The Umbrella-Upgrading Hypothesis)

Summary of the New Umbrella Model of Online Newspaper Geography
The new umbrella model can be summarized as such: Technologically, the
Internet's global nature enables online newspapers to serve local and
long-distance audiences. The print edition transfers its market boundary to the
online edition. The transferred umbrella constitutes the local market, the size
of which is positively related to print circulation. The Internet also upgrades
the online market to the next higher geographic level, which, however, usually
is not the highest global level.
Methods
The rapid growth of the Internet population has been identified by various
sources. A national survey conducted by the Pew Research Center for the People
and the Press (1999) during November 1998 showed that 41 percent of the U.S.
population 18 years of age or older (75 million adults) used the Internet -- up
from 14 percent in 1995. The Pew Center found that the number of Americans who
go online to get news at least once a week has tripled during a three-year
period, starting from 4 percent in 1995 to between 15-26 percent in 1998 (1999).
An email survey of online newspapers was conducted to obtain information about
their geographic markets. Special attention has been given to the relationship
among three markets: the print edition's market, the online edition's local
market, and the online edition's long-distance market. Print circulation and
online usage data can be used to test the umbrella-transferring hypothesis (H1a)
and the umbrella-correlation hypothesis (H1b). These two hypotheses specify the
relationship between the print market and the online local market in terms of
market boundary and market size, respectively. Online usage data in the
long-distance market can test the umbrella-upgrading hypothesis (H2), which
deals with the impact of the Internet's global capacity on online newspapers'
market expansion. Online newspaper economics also was examined by inquiries
about current and plausible revenue models.
Definitions of Key Concepts
Several key concepts in the umbrella-transferring, umbrella-correlation, and
umbrella-upgrading hypotheses need to be defined. Newspaper directories and the
email survey can help measure these concepts.
Market Boundary & Market Size
A market is where business activities take place. For a print newspaper, market
boundary refers to the geographic border of its circulation area. Market size is
represented by circulation or penetration figures. Both can be measured by
obtaining circulation and commercial trading area information made available by
the Audit Bureau of Circulations and published in newspaper directories such as
Editor & Publisher International Yearbook, Working Press of the Nation, and
Bacons' Newspaper Directory.
For an online newspaper, market boundary is the border of the area where its
users reside. Market size is represented by online usage. However, practically,
online audience measurement remains a difficult problem facing most Web sites.
The exact number of online users often is a mystery because no single reliable
technique exists to accurately measure the size of audiences and their identity
(Greenstein, 1998; Stone, 1998a; Stone, 1998b; Stone, 1998c; White paper: How
interactive ads, 1998; Kirsner, 1997; Gugel, 1997; Rich, 1997). Conference with
six online professionals at newspapers of different sizes about this matter
suggested that most online newspapers should be able to provide "page views" and
"unique users" statistics. "Page views" (or "impressions") -- the number of
requests for HTML pages -- is accepted by online advertisers as a standard
measure of online audience size. "Unique users" -- the number of site visitors
-- is another commonly used audience measure, often derived by the use of
"cookies" (a text file a server places in a visitor's computer). Both measures
provide one dimension of the online market size and can serve comparison
purposes. Therefore, the market size concept is measured by asking respondents
to report page views and unique users data.
The online market boundary is measured by asking respondents to estimate the
percentages of online users who reside within/outside the print edition's
primary circulation area. Respondents were asked to specify their measurement
methods for more accuracy.
Since online newspapers also can intentionally define their market boundaries by
targeting audiences in certain geographic areas, the online market boundary also
was measured by "the intention to target audiences in a certain geographic
area."
Local/Long-Distance Markets
An online newspaper's local market is where both online and print editions are
available. An online newspaper's long-distance market is where only the online
edition is available. Practically, the existence of local/long-distance markets
depends on substantial usage from within/outside the print edition's circulation
area.
Methodological Justification
Traditional media markets are monitored by institutions such as Nielsen (for
television) and the Audit Bureau of Circulations (for newspapers), and market
data often are available to the public. Some companies specializing in online
audience measurement -- such as RelevantKnowledge, MediaMetrix, I/Pro, @Plan --
have been trying to serve as the equivalent of ABC for the Internet. The ABC
also offers an auditing service for Web sites called ABC Interactive. But there
still is no single source where usage statistics for online newspapers are made
available. Therefore, a survey has to be conducted to gather baseline data.
Since most online newspapers provide an email link on their Web sites (Tremayne,
1998; Gubman & Greer, 1997), one of the most efficient ways to reach these
online news providers is via email.
For email surveys, attaining a decent response rate is problematic. Previous
studies surveying online news organizations or journalists via email showed
response rates ranging from 71 percent (Harper, 1996), 35 percent (Alexander,
1997), 32 percent (Tankard & Ban, 1998), to 13 percent (Ban, 1998). Because the
primary purpose of this study is to identify the relationship between a
newspaper's print and online markets -- providing descriptive statistics about
the population is secondary -- a moderate response rate around 30 percent should
be acceptable.
Sampling
Since a market, as represented by the umbrella symbol, is a geography-sensitive
concept, the sample must include newspapers operating at each level of the
hierarchy -- national newspapers, metropolitan dailies, and community
newspapers. There are 1,565 dailies and 8,858 weeklies in the U.S. (Bacon's
Newspaper Directory, 1997). Due to the large number of newspapers in the U.S.
and to time constraints, sampling is necessary. To increase generalizability,
stratified random sampling on a national scale is an option. But from such a
large population, it is difficult to get a sample that can illustrate a complete
umbrella structure. Instead, choosing a few states and conducting a census
within each may yield a better portfolio of online newspapers. So, four states
-- California, Texas, New York, and Florida -- were selected for a number of
reasons. First, with California in the West Coast, Texas in the South, New York
in the Northeast and Florida in the Southeast, a four-state sample yields more
representative results than a one-state sample. Secondly, these are the states
in the U.S. with the largest populations. Together these four states accounted
for more than 30 percent of the U.S. population (U.S. Census Bureau, 1999).
California, New York, Florida, and Texas also were identified by Nielsen Media
Research as the top four states with the largest number of Internet users. Each
was estimated to have at least two million online users in fall 1997
(CommerceNet/Nielsen, 1998). In addition, each of these states has a substantial
number of newspapers (Editor & Publisher International Yearbook, 1998). These
characteristics increase the representativeness of the study.
Since only a handful of U.S. daily newspapers operate on a national scale, all
were included in the sample. Online newspapers in California, Texas, New York,
and Florida were taken from the Editor & Publisher's online media directory at
www.mediainfo.com (1998) and the Newspaper Association of America's online
newspaper portal at www.newspaperlinks.com (1998). Included were
English-language newspaper sites currently offering general-interest news
content. For those that offered an email directory of online staff, the survey
was sent directly to the person in charge of marketing matters. For sites with
only a general email address, the survey was sent to that address. A few sites
offered no email address and thus were excluded from the study. The final sample
consisted of 193 online newspapers, of which four operate at the national level,
65 in California, 52 in Texas, 38 in New York, and 34 in Florida.
Survey Instrument
The draft of the email questionnaire was revised several times based on
suggestions made by six online newspaper professionals to make sure the
questions were answerable and relevant from the industry's point of view. The
survey was then pre-tested on 13 online newspapers. The email survey consists of
a cover letter and 11 questions about online newspaper market definitions and
revenue models. Whenever the target respondent was identifiable, the cover
letter was directly addressed to that person because a personalized message may
increase the response rate. In other cases, the recipient was asked to forward
the survey to the marketing practitioner or, if there was no one in marketing,
the most appropriate person who could help answer the questions.
The 11 questions, which took a person familiar with usage data about 5 minutes
to complete, focused on site traffic measures, geographic markets, target
audiences, and revenue models. (Appendix).
The size of the online audience was measured by asking:
[    ] Estimated number of unique online USERS (not hits) per DAY
[    ] Estimated number of page views per MONTH
The breakdown of the local and long-distance markets was measured by asking:
Where do you think your audience resides?
[    ] % reside WITHIN the print edition's primary circulation area
[    ] % reside OUTSIDE the print edition's primary circulation area
[    ] Don't know
The intention to target audiences in a certain geographic area was measured by
this question:
Your target audience is (Check all that apply)
[  ] WITHIN the primary circulation area of the print edition
[  ] OUTSIDE the primary circulation area of the print edition. If so,
[  ] in what area?

Data Collection
The survey of 193 online newspapers -- including four national dailies and those
in California, Texas, New York, and Florida -- was conducted via email and fax
between Dec. 17, 1998 and Feb. 19, 1999. Each newspaper received one mailing and
up to two follow-up surveys.
Completed surveys were received from 64 out of the 193 online newspapers. A few
online newspapers, instead of answering survey questions, responded by saying
that company policy did not allowed them to disclose usage and revenue
information to people outside their organization. Some said their Web site
functioned as a public service and thus the survey, which focuses on market
issues, was irrelevant to them.
The response rate for this survey can be estimated in two ways. The gross
response rate is 41 percent, calculated by dividing the number of replies
(including non-completion responses and completed surveys) by sample size. The
completion rate is 33 percent, calculated by dividing the number of completed
surveys by sample size. The completion rate yielded a sampling error of 8.2
percent.[4]

Major Findings
Responses were received from 52 daily and 12 weekly newspaper sites. The size of
the online staff -- including editors, sales, and technicians -- ranged from 0
to 130. ("Zero staff members" probably means there is no full-time person in
charge of the online product). The mean was 12.5, the median was 3, and the mode
was 2. (Table ).

Table 1:        Size of Online Staff
Number of Employees
%
100 -130
5
11 - 50
22
5 - 10
11
3 - 4.5
13
2 - 2.5
21
1 - 1.5
16
Less than 1
13


(Valid cases)
(63)
Mean = 12.5; Median = 3; Mode = 2.

Circulation of the print edition ranged from 693 to more than 1.5 million (Table
2), based on which, newspapers (print editions) were placed into three
categories: community (circulation under 100,000), metro (circulation between
100,001 and 1 million), and national (circulation over 1 million and those
listed as national newspapers in Editor & Publisher International Yearbook).
Five national newspapers, 12 metro newspapers, and 47 community newspapers
participated in this study.
Table 2:        Circulation of the Print Edition
Circulation
%
Over 500,000
8
100,001 - 500,000
16
50,001 - 100,000
18
25,001 - 50,000
16
10,001 - 25,000
20
0 - 10,000
21


(Valid cases)
63
Mean = 168,474; Median = 37,885.

To examine content-sharing between an a newspaper's print editions, the survey
asked respondents how much content in the online product is unique -- not
available in the print edition. Results showed larger newspapers tended to
produce more unique content than smaller ones, but, overall, the percentage of
unique online content was low. (Table 3).
Table 3:        Percentage of Unique Online Content

Community Newspapers
Metro Newspapers
National Newspapers
Overall
Mean
19
25
45
22
Median
10
20
35
10
Mode
0
20
N/A
0
Note: Cell entries are percentages.

Online Audience Measurement
Four Web sites did not reveal their page view statistics. Among the rest, the
number of page views per month ranged from 20 to 120 million. About one-fourth
(26%) of the sites had no more than 20,000 page views per month. About 22
percent had between 20,001 to 200,000; another 26 percent had between 200,001 to
1,000,000. About 16 percent had between 1,000,001 and 10,000,000 on a monthly
basis. Only 10 percent of them had page views over 10,000,000. (Table 4).

Table 4:        Page Views per Month
Page Views per Month
%
Over 10,000,000
10
1,000,001 - 10,000,000
16
200,001 - 1,000,000
26
100,001 - 200,000
12
20,001 - 100,000
10
0 - 20,000
26


(Valid cases)
(59)
Mean = 5.9 million; Median = 240,000.

Of the 64 online newspapers, 54 were able to provide usage data in terms of
daily unique users. The number of unique users per day ranged from 2 to one
million. The mean was 37,349, and the median was 2,300. About 40 percent of the
sites had no more than 1,000 unique users per day. About 20 percent had between
1,001 and 5,000. Some 27 percent had between 5,001 to 50,000. Another 12 percent
of them had more than 50,000 unique visitors per day. (Table 5).

Table 5:        Number of Unique Users per Day
Unique Users per Day
%
Over 50,000
12
5,001 - 50,000
27
1,001 - 5,000
21
0 - 1,000
40


(Valid cases)
(54)
Mean = 37,349; Median = 2,300.

Although there is no way to test rigorously how accurate these self-reports of
audience sizes are, knowing how these estimations were made might help. In
addition, asking the respondents to specify how they track site traffic can help
understand the methods currently used for online audience measurement. Results
showed that computer software was the most popular audience measurement tool
used by 65 percent of these online newspapers; WebTrends, RealMedia Open Ad
Stream, MKStats, and Analog were most commonly mentioned packages. About 30
percent received reports on site traffic from their parent companies. About 8
percent used market research data such as surveys. Another 8 percent used
auditing services. About 10 percent relied on feedback from individual users and
intuition. In general, online newspapers utilized a variety of measurement tools
to estimate their audience size.
Umbrella-Transferring
To test the umbrella-transferring hypothesis (H1a: An online newspaper's local
market boundary is defined by the print edition's market boundary.), the
relationship between the print edition's market boundary and the online
edition's local market boundary needed examination.
Of the 64 online newspapers surveyed, 53 reported having users residing within
the print edition's primary circulation area, including seven which didn't
provide exact percentages as requested. In other words, more than 80 percent of
online newspapers had users from within the print edition's circulation area.
Among these 53 online newspapers, 29 (55 percent) said more than 50 percent of
their online users resided within the print edition's circulation area. Of the
46 online newspapers reporting percentages, 63 percent said as much. (Table 6).

Table 6:        Percentages of Online Users Within the Print Edition's Primary
Circulation Area
Percentage of Online Users Within the Print Edition's Primary Circulation Area
Percent reporting this percentage
(all cases reporting)
Percent reporting this percentage
(cases reporting percentages only)
0 - 25 %
6
7
26 - 50 %
27
31
51 - 75 %
36
41
76 - 100 %
19
22
Percentages Unknown
13
N/A



(Valid cases)
(53)
(46)
Mean = 62; Median = 66. (Cases reporting percentages only.)

Market boundaries also were measured by the intention of targeting a certain
geographic market. Almost 90 percent of respondents said they targeted audiences
within the primary circulation area of the print edition. Among them, 74 percent
targeted readers and non-readers of the print edition, 18 percent targeted
non-readers only, and 7 percent exclusively targeted print readers only.
Both measures of online market boundaries (geographic distribution of users, and
targeting intention) serve as a strong indication of market-transferring between
the print and the online editions. In other words, analysis supports the
umbrella-transferring hypothesis: An online newspaper's local market boundary is
defined by the print edition's market boundary.
Umbrella-Correlation
To test the umbrella-correlation hypothesis, the size of the print market and
the size for the online local market had to be determined. The print edition's
market size is represented by circulation. The online edition's local market
size can be calculated using an indirect measure -- overall market size
multiplied by percentage of online users within the print edition's circulation
area. The online edition's overall market size was measured twice -- in terms of
page views (Table 4) and in terms of unique online users (Table 5). The
percentages of online users within the print edition's primary circulation area
also were known. (Table 6). Therefore, the size for the online edition's local
market in terms of page views and unique users was determined, respectively.
Correlation analysis was utilized to identify the relationship between the print
market and the online product's local market. Because both variables are on a
ratio scale, Pearson's correlation coefficient (r) was calculated[5]. Results
showed that print circulation was positively related to page views in the online
local market (r = .8263, p = .000) and unique users in the online local market
(r = .6309, p = .000). In other words, a substantial relationship between the
size of the print market and the size of the online local market supports the
umbrella-correlation hypothesis (H1b): The larger the print market is, the
larger the online product's local market is.
Umbrella-Upgrading
To test the umbrella-upgrading hypothesis (H2: An online newspaper's
long-distance market is one level above the print edition's market.), the
relationship between the print edition's market and the online edition's
long-distance market needed examination. Analysis focused on market boundary
because umbrella-upgrading basically is about market expansion across geographic
levels. The relationship between the two markets in terms of market size also
was explored.
In terms of market boundary, of the 64 online newspapers surveyed, 48 reported
having users residing outside the print edition's primary circulation area,
including two who didn't provide exact percentages as requested. In other words,
three-fourths of online newspapers had users from outside the print edition's
circulation area -- an indication of the Internet's boundary-transcending
capability. Among these 48 online newspapers, 64 percent reported that
long-distance usage accounted for more than 25 percent of total traffic. Of the
46 online newspapers reporting percentages, 67 percent said as much. (Table 7);
i.e., 50 percent of all respondents reported having a long-distance audience
that accounted for more than 25 percent of online readership.

Table 7:        Percentages of Online Users Outside the Print Edition's Primary
Circulation Area
Percentage of Online Users Outside the Print Edition's Primary Circulation Area
Percent reporting this percentage
(all cases reporting)
Percent reporting this percentage
(cases reporting percentages only)
0 - 25 %
31
33
26 - 50 %
50
52
51 - 75 %
10
11
76 - 100 %
4
4
Percentages Unknown
4
N/A



(Valid cases)
(48)
(46)
Mean = 38; Median = 34. (Cases reporting percentages only.)

In terms of online newspapers' intention to target long-distance audiences, 61
percent of respondents said they targeted audiences outside the print edition's
primary circulation area. To provide a larger picture, "intention to target
long-distance audiences" and "intention to target local audiences" should be
evaluated simultaneously. As noted earlier, 89 percent of online newspapers
target audiences within the print edition's primary circulation area. Comparison
showed that the majority (55%) of online newspapers targeted both local and
long-distance markets, 34 percent targeted the local market only, and 6 percent
targeted the long-distance market only. (Table 8).

Table 8:        Intention to Target Local and Long-Distance Audiences
Target Audience
%
Target Local and Long-Distance Audiences
55
Target Local Audience Only
34
Target Long-Distance Audience Only
6


(Valid cases)
64

In addition to the local/long-distance dichotomy, the umbrella-upgrading
hypothesis also specifies that the long-distance market is "one level above" the
print market. Therefore, a level-by-level analysis can help test the hypothesis.
Print markets were placed into three categories: community, metro, and national.
Online markets, on the other hand, were examined at the local, metro, regional,
national, or international level.
As noted earlier, Internet audience measurement still is problematic in terms of
accuracy. Practically, it is not feasible to collect data about the breakdown of
"online usage" at different geographic levels with a survey. Therefore, the "one
level above" part of the umbrella-upgrading hypothesis was tested using online
newspapers' "targeting intention." Respondents who targeted long-distance
audiences were asked to describe the geographic area where long-distance users
resided, and 27 (including 16 community newspapers, eight metro newspapers, and
three national newspapers) answered this open-ended question. Responses were
content analyzed and tabulated in Table 9. Results were mixed. All national
newspapers unanimously defined their long-distance market as international or
global -- one level above the print market, as predicted by the
umbrella-upgrading hypothesis. One-fourth of metro newspapers targeted a
long-distance audience at the regional level -- also one level above the print
edition's market. But another one-fourth of metro newspapers targeted the
international level. One-fifth of community newspapers also targeted the
international level and 13 percent at the regional level. The "one level above"
part of the umbrella-upgrading hypothesis was supported among all national
newspapers, some metro newspapers, and few community newspapers.

Table 9:        Geographic Level of the Long-Distance Online Market by Geographic Level
of the Print Edition

Geographic Level of Print Edition
(Cell entries are percentages)

Community
Metro
National
Geographic Level of Online Long-Distance Market
Community
0
0
0

Metro
6*
13
0

Regional
13
25*
0

National
6
13
0

International
20
25
100*

Psychicspace / Other**
56
25
0





(Valid cases)
16
8
3

*Note: These are long-distance market levels predicted by the umbrella-upgrading
hypothesis. (Community  Metro; Metro  Regional; National  International).

**Note: The "Psychicspace or other" category included responses with no
references to specific geographic areas, e.g. "former residents regardless of
place," "people who are going to move to this area," "geography is irrelevant on
the Web," etc.

To explore the relationship between the print market and the online
long-distance market in terms of market size, correlation analysis was
performed. Print circulation was positively related to page views in the
long-distance market (Pearson's r = .7591, p = .000) and unique users in the
long-distance market (Pearson's r = .8439, p = .000). The results implied that
umbrella-correlation also occurs between the print market and the online
long-distance market. In other words, the larger the print market is, the larger
the online product's long-distance market is.
Overall, it is evident that the Internet enables online newspapers to expand
their geographic market outside the print edition's circulation area both in
terms of actual usage and targeting intention. Many online newspapers
practically target and serve multiple audiences in multiple markets at multiple
geographic levels. As to whether the long-distance market is "one level above"
the print edition's market, data did not fully support this part of the
umbrella-upgrading hypothesis.
Profitability and Revenue Models
To examine the relationship between market size and economic viability -- a
major concern in the online newspaper industry, the survey included questions
about profitability and current/expected revenue sources. Respondents were asked
whether their online newspaper was making a profit. About 27 percent said yes,
while the majority (73%) said no. Although online newspapers with larger
audiences seemed more likely to make a profit, no relationship between
profitability and market size (of the print market, the online market, the
online local market, and the online long-distance market) was statistically
significant.
Respondents were asked to specify their current revenue sources. Almost 80
percent of respondents said online advertising was a revenue source. About 40
percent generated revenues by creating Web sites for clients. Classified
advertising (which was meant to be part of the "advertising" category) was
specified by as many as 18 percent of respondents in the "other" category -- an
indication of the great potential of online classified ads. E-commerce, the ISP
business, and pay-per-use services also generated revenues for more than 15
percent of online newspapers, while subscription was ranked the lowest among all
-- only 3 percent of the online newspapers surveyed charged users a subscription
fee. A couple also mentioned sponsorships as a revenue source in the "other"
category. (Table 10).

Table 10:       Current Revenue Sources
Revenue Models
Percent  Who Use Model
Advertising
79
Web Site Development
39
Classifieds*
18
E-Commerce
18
Serving as ISPs
16
Pay-per-Use Services
15
Subscription
3

* Advertising includes both banner ads and classified ads, but 18 percent of
respondents specified classifieds separately.

Since the online environment quickly changes, plausible models which may play a
role in the future also are of interest. Asked to evaluate the likelihood of
using these revenue models -- subscription, advertising, pay-per-use services,
Web site development, ISP, E-Commerce -- on a 5-point scale, almost all the
respondents agreed that online advertising is "likely" or "very likely" to serve
as an important revenue source for their online newspapers within the next two
years. E-commerce, and Web site creation also were rated as "likely" or "very
likely" by over 50 percent of the online newspapers. (Table 11).

Table 11:       Evaluation of Plausible Revenue Models in Two Years

Revenue Models
%

Very likely
Likely
Neutral
Unlikely
Very unlikely

Rank
Advertising
88
7
2
0
0
1
E-Commerce
35
29
15
7
11
2
Web Site Development
33
35
13
0
17
3
Pay-per-Use Services
27
21
21
10
16
4
ISP
10
0
14
18
59
5
Subscription
4
2
14
15
65
6
Note: Ranks were calculated based on the average scores assigned by respondents
using the 5-point scale measures.

When comparing the current and future revenue models specified by these online
newspapers, the advertising model dominated and probably will continue to do so
in the near future. On the other hand, E-commerce, although being used by only
18 percent of the sites, was expected to play a more important role in the near
future.
Summary & Discussion
Of the 64 online newspapers surveyed, while five percent had more than 100
employees, the median staff size was three. While 10 percent had over 10 million
page views per month, the median was 240,000. While 12 percent had more than
50,000 unique visitors per day, the median was 2,300.
Staff size, page view, and unique user statistics seemed to show that the online
newspaper industry resembled a "bi-modal market structure" -- on one tier is an
oligopoly (3-4 large firms) that controls 70-90 percent of the market, while on
the other tier a large set of small firms fighting for remaining shares
(Albarran, 1998). According to Albarran (1998), this new type of market
structure, already seen in the cable television and ISP industries in the U.S.,
will replace many existing market structures especially in the new media
marketplace.
The umbrella-transferring hypothesis (H1a) -- which specifies the relationship
between an online newspaper's local market boundary and the print edition's
market boundary -- was fully supported. Almost 90 percent of online newspapers
targeted audiences within the print edition's primary circulation area. More
than 80 percent of online newspapers had users who resided within the print
edition's market, of which 55 percent estimated that the local audience
accounted for than 50 percent of total traffic. Therefore, an online newspaper's
local market boundary is defined by the print edition's market boundary.
The umbrella-correlation hypothesis (H1b) -- which specifies the relationship
between the print market and the online local market in terms of market size --
also was supported. In addition, umbrella-correlation also occurs between the
print market and the online long-distance market. Correlation analysis showed
that print circulation is highly correlated with the online market size -- for
local (r = .8263 by page views, r = .6309 by unique users) and long-distance (r
= .7591 by page views, r = .8439 by unique users). In other words, the larger
the print market is, the larger the online market is.
The umbrella-upgrading hypothesis -- which identifies the existence and the
location of online newspapers' long-distance market -- was partly supported.
More than 60 percent of online newspapers targeted audiences outside the print
market and 50 percent reported having more than 25 percent of online readership
from outside the print edition's circulation area. Both measures of online
market boundaries (geographic distribution of users, and targeting intention)
indicated the existence of a long-distance market. Furthermore, 55 percent of
online newspapers simultaneously targeted local and long-distance audiences,
seeking markets at multiple geographic levels. It is evident that the Internet
has the market-upgrading capacity that expands the online market up from the
print market to a higher geographic level.
As to whether the long-distance market is "one level above" the print market,
results were mixed. The argument was supported among all national newspapers,
some metro newspapers, and few community newspapers. Several reasons may explain
why this "one level above" rule was not fully supported. First, when an online
newspapers has a specific niche (e.g. San Jose Mercury Center's coverage of
high-tech industries in Silicon Valley may have great national and international
appeal), geographic proximity alone can no longer explain the geographic
distribution of the online audience. Second, asked to describe where their
target long-distance audience resides, many respondents didn't make references
to specific geographic areas but mentioned that their long-distance audiences
were former or future residents (those who are going to move to the area). The
geographic distribution of these people probably also is to a certain extent
subject to proximity. (e.g., former or future residents of Houston are more
likely to be found in Texas than in Pennsylvania or in Europe.) Finally,
analysis relied on the 27 self-reports on their target long-distance market
because online usage data for multiple geographic levels were not available.
However, targeting intention may not reflect actual usage. Therefore, the "one
level above" part of the umbrella-upgrading hypothesis should be re-examined
when more specific usage data are available.
Although results showed that the local market is twice as important as the
long-distance market in terms of online usage -- on average, online newspapers
reported about two-thirds of audiences within (62%) and one-third (38%) outside
the print edition's circulation area -- this may simply indicate that the
potential of a long-distance market has not been fully developed. Online
publishers should consider going beyond the market defined by the print edition
to further cultivate the long-distance market.
Overall, empirical evidences have demonstrated that the new umbrella model
effectively captures the market relationship between online and print
newspapers.
The survey also provides valuable information about profitability and revenue
models. No relationship between market size and profitability was statistically
significant. An explanation is that larger newspapers (in print) tend to attain
a larger online audience, which may bring more revenues. However, larger
newspapers also tend to devote more resources to their online product -- as
evidenced by higher percentage of unique content -- and thus are less likely to
be profitable in the short term. Therefore, market size probably is more related
to revenue rather than profit, but this study didn't examine revenue and cost
figures separately.
Most online newspapers experiment with different revenue models and hope to
become an economically viable medium. Advertising and E-commerce were expected
to serve as primary revenue drivers in the near future, while the subscription
model probably won't work for the online medium.
Since most revenue sources are traffic-based -- revenues increase as the number
of online users increases - theoretically, market size should differentiate
online newspapers' economic viability. As to what determines an online
newspaper's market size, umbrella-correlation demonstrated a significant
relationship between print circulation and online market size.
To take a closer look at print circulation, in this study, it ranged from less
than 1,000 to more than 1.5 million. It explains why online newspapers, despite
being equally accessible on the Web, differ greatly in terms of audience size.
In other words, the print market greatly affects the online edition's market
size and market boundary.
Therefore, it is unrealistic to think about an online newspaper without taking
into account the role of its print edition in a marketplace. Although some sites
denied being categorized as an online "newspaper," at this stage, the market
relationship between the print and online editions is too influential to be
ignored. In other words, the print edition's market size would affect the online
newspaper's market size, which, in turn, would affect an online newspaper's
revenue-making capability.
Factors such as population and competition affect print newspaper circulation.
These factors also, in turn, affect an online newspaper's market size. For
example, online news sites targeting audiences in large cities have better
chances to attain a larger online audience than those in smaller cities.
In addition to the nature of the print newspaper, nature of the audience and
nature of the online product also affect an online newspaper's market size (Chyi
& Sylvie, 1998).
Conclusions
Geography is not irrelevant in cyberspace. Although the online newspaper is an
inherently global medium, it still has market boundaries. Online newspapers
still are a local product for all intents and purposes, despite the Internet's
boundary-spanning capabilities. The ramifications shouldn't be lost on most
media managers and, to a lesser degree, on media scholars.
The former should beware that -- unless they are employed at one of the larger
(possibly national) newspapers -- even if their online long-distance audience is
substantial, the product still is local because the print edition largely
defines the online local market boundary. The consequences lie in the issue of
substitutability; i.e., how substitutable is an electronic product that -- as
this study's data suggest -- essentially mimics the print product? Obviously,
the more substitutable, the less should be the intent to target readers of the
print product. But given the fact that the papers in the sample averaged 22
percent of unique online content, one would expect more would say they intend to
target long-distance audiences. However, almost 90 percent said they targeted
local audiences, of which 81 percent targeted readers of the print edition -- an
indication of a heavy focus on local audiences. In contrast, only 61 percent
targeted long-distance audiences. This may be a reflection of lack of resources
and profits or, at the very least, an emphasis (on the part of the newspaper) on
attracting new readers via the Internet. Still, it presents cause for concern
because it assumes people who don't read the newspaper will, instead, read it in
electronic form.
On a theoretical level, this clarifies earlier suggestions regarding product
differentiation (Chyi and Sylvie, 1998). Conventional economic wisdom says
online newspapers must show their products can adequately complement traditional
media and other online offerings, "to develop distinctive content and to draw a
distinctive audience that is as homogeneous as possible" (p.16). And yet, as
this study shows, that's exactly what newspapers are not doing. In fact, they
seem to be doing the opposite. Whether this is a short-term phenomenon prompted
by lack of profits or by ignorance remains to be seen, but the lesson for
scholars may be that profits are not the barometer for success at this time.
This study's data obviously place advertising as the revenue model of the
future, indicating profits are still in the picture, but maybe not now and maybe
not at the local level. Additional study should examine the stage of development
of online newspapers who insist on maintaining less-than-unique content to
determine whether any diffusion principles (regarding clarifying and routinizing
procedures of adoption; see Rogers, 1983) may be present.
If the profit motive continues to be the goal, this study's findings call into
question the economic, long-term viability of the local electronic newspaper
operation bent on relying on shovelware for content - unless advertising
strategies are changed or market demographics and technology usage alter their
present course substantially. The finding that the print market defines the
online market indicates that local advertisers would only be attracted to a
less-than-unique product if it delivered an audience that is new or unique,
i.e., one that is not similar to one delivered by the print product. The
likelihood of the Internet and the printed product delivering the same type of
audience is potentially small, but then so is the likelihood that the
Internet-savvy audience would be attracted to the online newspaper. So that
means electronic newspapers wouldn't deliver attractive local audiences in
sufficient numbers. The key then, for online newspapers, is to market that
less-than-unique local product to an audience that would find it unique: a
long-distance audience, one in "psychicspace." This strategy would make the
smaller online operations more likely to be profitable because they could match
the long-distance audience to appropriate, non-local advertisers. In short,
newspaper executives must stop thinking "local" when it comes to online markets
because such thinking short-circuit's the medium's advertising potential. As
this study shows, the larger the print market, the larger the online product's
long-distance market. And as the umbrella opens, profitable newspaper operations
must follow.

Appendix: Email Survey Questionnaire
Subject: 5 minutes in exchange for market research

Dear online journalist at [name of the online newspaper]:

We are conducting a research project concerning online newspaper economics. We
would appreciate it if you could help us complete this brief survey which should
take no more than 5 minutes. If you would like to receive a report of our
findings or related studies, please let us know by marking the last question.

We would like to reach the person in charge of MARKETING matters in your
organization. If you are not that person, please forward this to the most
appropriate person. Thank you, in advance, for your assistance.

Your organization was selected from www.newspaperlinks.com and the Editor &
Publisher Directory. Your individual responses will be kept confidential.

Directions: To answer a question, type an x between the brackets: [ x ]. For
open-ended questions, type your answers between brackets: [ your response ].
Please make no other changes to this survey. Thank you.

Sincerely,
H. Iris Chyi / George Sylvie
Ph.D. Candidate / Associate Professor
Department of Journalism
The University of Texas at Austin
FAX: (512) 471-7979 (if you prefer to respond by FAX)

(Authentication marker-- &F^%06*%670**01&0101-- do not remove.)

[    ] 1. How many people are employed in your online news operation (including
editors, sales, and techs)?

[    ] 2a. Estimated number of unique online USERS (not hits) per DAY

[    ] 2b. Estimated number of page views per MONTH

2c. Your estimation is based on: (Check ALL that apply)
[    ] Computer software, which one(s) [    ]
[    ] Report from parent company
[    ] Market research data (e.g. surveys)
[    ] Anecdote and intuition
[    ] Audit by [    ]
[    ] Other, please specify [    ]

3. Where do you think your audience resides?
[    ] % reside WITHIN the print edition's primary circulation area
[    ] % reside OUTSIDE the print edition's primary circulation area
[    ] Don't know

4. Your target audience is (Check ALL that apply)
[    ] WITHIN the primary circulation area of the print edition.
[    ] OUTSIDE the primary circulation area of the print edition. If so,
[    ] in what area?

5. Your target audience is (Check ALL that apply)
[    ] Readers of the print edition
[    ] Non-readers of the print edition
[    ] Other, please specify

6. [    ] % of the online non-advertising content is unique (i.e. NOT provided
by the print edition associated with your site.)

7. Which of the following is (are) NOW serving as revenue source(s) for your
online newspaper? (Please provide the percentages.)
[    ] % of revenues from charging users a subscription fee
[    ] % from online advertising
[    ] % from pay-per-use services (e.g. news archive)
[    ] % from creating Web sites for clients
[    ] % from serving as an ISP (Internet service provider)
[    ] % from online shopping/E-Commerce
[    ] % Other source, please specify [    ]

8. Which of the following may serve as an important revenue source for your
online newspaper within NEXT TWO YEARS? Please use the five-point scale for each
to estimate how likely they are: 1) Very likely 2) Likely 3) Neutral 4) Unlikely
5) Very unlikely
[    ] Charging users a subscription fee
[    ] Advertising
[    ] Pay-per-use services (e.g. news archive)
[    ] Creating Web sites for clients
[    ] Serving as an ISP (Internet service provider)
[    ] Online shopping/E-Commerce
[    ] Other source, please specify [    ]

9. Is your online newspaper making a profit?
[    ] Yes
[    ] No

10. Please state the mission of your online newspaper?
[    ]

11a. To show our appreciation for your participation, we would be happy to share
our research findings with you. Please let us know if you are interested in
getting
[    ] results of this study
[    ] information about related studies we've conducted

11b. How can we reach you?
[    ] Your personal email address
[    ] Your name and job title

Thank you very much for your participation!
HIDDEN TEXT: Optional-must be placed in this order if it is included in the
dissertation. If you don't want to include a glossary, then delete the entire
page and the following page break.
 References
Albarran, A. B. (1998). Convergence in the mass media industries: Implications
for the marketplace and the university curriculum. Paper presented to the Media
Management Division at the AEJMC Mid-Winter Meeting, February 21, 1998, Dallas,
Texas.
HIDDEN TEXT: A reference/bibliography section must be included at the end of
your dissertation, even if you also use chapter end notes. Any standardized
format for a reference or bibliography list is acceptable IF it is approved by
your supervising professor and if you follow it consistently.
Alexander, M. J. (1997). Net gain? New England's online newspapers assess
benefits and drawbacks of their electronic editions. Paper presented to AEJMC,
Chicago.
Bacon's Newspaper Directory. (1997). Chicago: Bacon's.
Ban, H. (1998, February 21). Online Journalists: Their Role Perception and Job
Satisfaction. Paper presented to the AEJMC Mid-Winter Conference on Media
Convergence, Southern Methodist University, Dallas, Texas.
Chyi, H. I. & Sylvie, G. (1998). Competing with whom? Where? And how? A
Structural Analysis of the Electronic Newspaper Market. The Journal of Media
Economics, 11(2), 1-18.
CommerceNet/Nielsen media research. (1998, June 10). Internet states. [Online].
Available: http://www.nielsenmedia.com/interactive/commercenet/F97/br3b.htm
Editor & Publisher International Yearbook. (1998). New York: Editor & Publisher.
Editor & Publisher Interactive. (1999, March 20). MediaInfo links: Online media
directory. [Online]. Available:
http://www.mediainfo.com/ephome/npaper/nphtm/statistics.htm
Fitzgerald, M. (1997, February 16). 68% of news Web sites generate own content:
Results of new survey announced at conference. Editor & Publisher Interaction
Conference Daily News [Online]. Available:
http://www.mediainfo.com/ephome/Interactive97/stories/021597c3.html
Greenstein, J. (1998, November). Clickthrough: How many? How much? Who knows?
Bill's Content, 1(4), 54-56, 58.
Gubman, J. & Greer, J. (1997). An analysis of online sites produced by U.S.
newspapers: Are the critics right? Paper presented to the AEJMC, Chicago.
Gugel, C. (1997). Tracking the Web: Measuring results from ad exposure to ROI.
Agency, 6 (4), 42-43.
Harper, C. (1996). Online newspapers: Going somewhere or going nowhere?
Newspaper Research Journal, 17(3-4), 2-13.
Kirsner, S. (1997). Web of confusion. American Journalism Review, July/August
1997, 34-39.
Lacy, S. & Simon, T. F. (1993). The Economics & Regulations of U.S. Newspapers.
Norwood: Ablex.
Newspaper Association of America. (1998, Nov. 15). Online newspaper listings.
[Online]. Available: http://www.newspaperlinks.com
Newspaper Association of America. (1999, March 20). Facts about Newspapers: A
statistical summary of the newspaper industry. [Online]. Available:
http://www.naa.org/info/facts/
Picard, R.G. (1989). Media economics: Concepts and issues. Newbury Park, CA:
Sage.
Poindexter, P. M. & McCombs, M. E. (1998). Making Research Relevant.
(Forthcoming). Bedford/St. Martin's.
Rich, L. (1997, February 3). Count them in. Adweek's Interactive Quarterly,
24-29.
Rosse, J. N. (1975). Economic limits of press responsibility. Studies in
Industry Economics, 56. Stanford, CA: Stanford University, Department of
Economics.
Stone, M. L. (1998a, September). Don't believe the numbers: Web traffic stats
remain unreliable and misused. Mediainfo.com, 4-6, 8.
Stone, M. L. (1998b, September). Nasty battle over Web ratings goes public: Top
web sites argue over who has the most readers. Mediainfo.com, 10, 12-13.
Stone, M. L. (1998c, September). Chasing a holy grail: What does "Reach" really
mean? Mediainfo.com, 14.
Tankard, J. W. & Ban, H. (1998, February 21). Online Newspapers: Living Up to
Their Potential? Paper presented to the AEJMC Mid-Winter Conference on Media
Convergence, Southern Methodist University, Dallas, Texas.
Tremayne, M. W. (1998). The Transforming Potential of the Internet: An Analysis
of Online News and the Use of Interactivity and Nonlinear Storytelling. Master's
thesis. The University of Texas at Austin.
U.S. Census Bureau. (1999). 1997 Population Estimates. [Online]. Available:
http://www.census.gov/
White paper: How interactive ads are delivered and measurement implication.
(1998, July 1). Audit Bureau of Circulations Interactive. [Online]. Available:
http://www.accessabvs.com/webaudit/admeasurement.html
Working Press of the Nation. Vol. 1: Newspapers. (1999). New Providence, NJ:
National Register Publishing.

[1]  Includes Web sites and partnerships with consumer online companies
[2]  Includes sites on BBS, Microsoft Network, AOL, Prodigy and Gopher services
[3]  About 90 percent of online newspapers provided local news (Gubman & Greer,
1997).
[4]  A correction factor for smaller populations (1 - n/N) was introduced when
calculating the sampling error.
[5]  Pearson's correlation coefficient (r) measures the magnitude of the linear
association between two variables (Poindexter & McCombs, 1998). It ranges from
negative one (-1) to plus one (1), where zero (0) means the absence of a linear
relationship, one (1) means a perfect positive linear relationship and negative
one (-1) means a perfect negative relationship.