This paper was presented at the Association for Education in Journalism
and Mass Communication in Toronto, Canada, August 2004.
If you have questions about this paper, please contact the author
directly. If you have questions about the archives, email
[log in to unmask] For an explanation of the subject line, send email to
[log in to unmask] with just the four words, "get help info aejmc," in the
body (drop the "").
(Oct 2004)
Thank you.
Elliott Parker
************************************************************************
Ordinary People and the Weekly Newspaper:
The Influence of Community Diversity on Source Diversity
John Hatcher
Ph.D. Student
S.I. Newhouse School for Public Communication
Syracuse Univesrity
31 Perry Place
Canandaigua, NY 14424
(585) 394-6496
[log in to unmask]
Please consider for the MacDougall Student Paper Award.
Submitted to the Newspaper Division of the Association for Education in
Journalism and Mass Communication.
Abstract
It has been said that community journalism, small daily and weekly
newspapers, report on the "everyday lives of ordinary people," suggesting a
uniformity of content based on routinized practices. However, research
suggests community heterogeneity has a strong influence on the content of
newspapers. This study asks whether community type can be used to predict
the variability in source diversity in weekly newspapers. This study uses a
content analysis of 659 sources from 40 issues of 10 different weekly
newspapers to ascertain the reliance on official sources in reporting the
news. The results find partial support for the prediction that community
diversity influences source diversity in weekly newspapers.
Ordinary People and Weekly Newspapers:
The Influence of Community Diversity on Source Diversity
It has been said that community journalism reports on the "everyday lives
of ordinary people" (Lauterer, 2000, p. xxii), but how do community
newspapers differ from one another? Are journalists at community
newspapers, small daily and weekly newspapers, governed by a set of
routines that will mean a uniformity of content and source selection
regardless of the community in which they exist? Or will the influences of
different types of communities mean that this unique population of
newspapers will have variations in content?
This paper explores differences in the use of official sources at weekly
newspapers looking at whether their content appears to be influenced by the
diversity of the community being served. Weekly newspapers circulating in
more pluralistic communities are predicted to rely more on unofficial
sources to report the news. Weekly newspapers with comparable circulation
sizes that serve communities with greater homogeneity, conversely, are
expected to include more voices of official sources. This paper reviews
existing literature exploring the subjects of source diversity, the
community integration theory, news framing, the guard dog perspective, and
other theories that attempt to explain the role of media in a community. A
content analysis of 659 sources from 40 issues of 10 different weekly
newspapers in New York was conducted to ascertain the reliance on official
sources in reporting the news and determine if the use of official sources
is dependent upon the heterogeneity of the readership area.
Theory
The use of "everyday" people appears to be a valued trait among journalists
in the community press. Community journalism is a term with myriad
definitions based on who is using it and in what context. For some, it
connotes images of the small-town newspaper run by one person who may write
the stories, take the pictures and even deliver the finished product. While
this example is certainly one part of the community press, the complete
definition is more diverse (Lauterer, 2000). Community journalism also
includes ethnic newspapers, faith-based publications, and newspapers
serving small "communities" that are part of larger metropolitan areas.
Loosely defined, community journalism shares a common denominator outlined
by former newspaper editor Vicki Simons: "It chronicles and comments on the
everyday topics that shape peoples' lives in our neighborhoods and in our
towns; it involves reporting on all manner of personal tragedies: fires,
auto accidents, crimes, children fighting ghastly diseases" (Lauterer,
2000, p. 7). Even in our nation's most densely populated cities, community
newspapers help form the identity of smaller neighborhoods. A publisher of
a community newspaper in the Bronx, N.Y., Pulitzer Prize-winner Bernard
Stein, said the job of the community journalist is "to cover the everyday
lives of ordinary people" (Lauterer, 2000, p. xxii). Taking these two
definitions into account, it seems journalists in the community press
strive to provide their readers stories that include the voices of people
just like them. Whether it's the results of the Little League baseball game
or coverage of neighbors in heated debate at the village board meeting, the
community press doesn't view any voice as insignificant. But does this
ideal reflect itself in the pages of community newspapers? This study will
explore this question. Going a step further, this study also will look at
community newspapers serving communities ranging from rural to metropolitan
in hopes of discovering whether the influences of the official voices can
be predicted by differences in the population.
The average circulation of community newspapers nearly tripled between 1965
and 1998 ('99 Facts about Newspapers, 1999), increasing the need for
scholarly work on this aspect of the nation's print media. There has been
much research on the relationship between journalists and sources in
framing the news, and several of these projects examining this relationship
at larger, daily newspapers have called for additional research at smaller
newspapers (Powers & Fico, 1994).
Studies of source diversity
When Voakes, Kapfer, Kurpius and Chern (1996) conducted a content analysis
of newspaper articles to measure the correlation between the diversity of
content and the diversity of sources, they included in their research a
secondary hypothesis that predicted that a larger circulation newspaper
would have greater source and content diversity in its articles. Their
results find quite the opposite: "In fact, smaller newspapers have higher
overall diversity scores than larger newspapers. This is particularly
pronounced in source diversity, where there is a considerable difference
between the large newspaper mean score (.264) and the mean score of the
small papers (.331)" (p. 487). They speculate that this result is brought
about because the small newspapers they studied use more wire service and
syndicate articles. It is important to note that the small newspapers used
in Voakes' study are operationalized as two daily newspapers with
circulation size of less than 10,000. No weekly newspapers are included in
this study.
There are some fundamental differences between Voakes' study and this one.
The weekly newspapers analyzed in this study use locally generated articles
almost exclusively. Therefore, this research will test whether Voakes and
colleagues' speculation that articles generated by wire services and
syndicates are the factors that contribute to increased source diversity in
his research.
Just as community journalists say their desire is to reflect the voices of
an entire community, many larger, daily newspapers have made an effort to
get away from letting the voices of the official source dominate news
pages. This is reflected in the civic journalism movement, which encourages
journalists to let the agenda of the citizens in their community dictate
the content of their newspapers, rather than those of the political
official (Massey, 1998). However, Massey's content analysis finds that one
paper espousing civic journalism did not prove to have a profound
difference in the use of unofficial sources when compared with other
newspapers or even when compared with itself before it began its civic
journalism campaign.
There is also research that suggests a greater diversity in sources leads
to more thorough reporting. Hansen (1991) examined Pulitzer Prize-winning
newspaper articles and found the prize-winning articles rely less on
government sources than the typical daily front-page newspaper coverage.
Her work suggests that reporters often rely on tested routines and routine
sources that often dictate the sources journalists go to when gathering
news. She concludes that enterprise reporting does a better job of seeking
out the unofficial voices who "can add interest and realism to their
reports" (p. 481).
Factors affecting source selection
There has been much research theorizing on the factors that contribute to
journalists' choices in sources used to gather and report the news. Gaye
Tuchman identifies three factors that encourage journalists to choose
sources that represent some type of institution (Tuchman, 1978, p 93):
• Most individuals, as news sources, have an axe to grind. To be believed,
an individual must prove his or her reliability as a news source.
• Some individuals, such as committee heads, are in a position to know more
than other people in an organization. Although they may have an axe to
grind, their information is probably more "accurate" because they have more
"facts" at their disposal.
• Institutions and organizations have procedures designed to protect both
the institution and the people who come into contact with it. The
significance of either a statement or a "no comment" must be assessed
according to the newsworker's knowledge of institutional procedures.
To summarize, the desire to obtain facts viewed as "verifiable" push
journalists toward sources in some official capacity. The "facts" purported
by the unofficial may be just as accurate, but Tuchman asserts that they
are harder for a reporter to verify as being such.
Shoemaker and Mayfield (1987) hypothesize that a journalist's background
and orientation, social and institutional pressures, newsgathering
routines, and the control of the people in power all affect the content of
newspapers. However, they also note that news content can't be explained by
one theory or one dominating factor. Instead, they propose the interplay of
many factors with one powerful influence being the ideology of the funder
of the media source. The ideological interests of each media "financier"
have a profound impact ultimately on how all these other factors affect the
final media product.
Shoemaker and Reese (1996) diagram a hierarchical model that determines how
journalists choose to portray issues. They begin with the influences
society itself has on shaping the content. The focus narrows to include the
influences of outside sources such as the community itself, then the
influences of the media organization, followed by journalists' routines and
ultimately, the individual, whose own characteristics also play some part
in the predicting content.
While the decisions of an individual reporter may play a part in dictating
what stories are covered and which sources are accessed, more recent
research suggests it is the routines of a newsroom that may play the more
dominant role. The role of a journalist as a "gatekeeper" is amplified in a
study (Shoemaker, Eichholz, & Wrigley, 2001) that attempts to ascertain the
factors that lead to a reporter's decision of what constitutes news. Their
work reinforces the notion that routines of news work such as deadlines
play a major role in dictating what potential events are deemed newsworthy.
While reporters may often see themselves as writing stories for their
readers, it is also possible that reporters may see officials as the only
sources capable of providing the information they deem necessary to produce
a story of quality. In a study of what qualities reporters use to judge the
quality of a news article, Salomone and colleagues (1990) find many of the
standard elements required for a quality news article – such as accuracy,
assessment of a threat – would more than likely come from a traditional
news source or expert on the topic being addressed.
When journalists pursue a story, they often do so with a hypothesis in mind
as to the predicted outcome of their investigation (Stocking & LaMarca,
1990). These assumptions about a story may also play a part in dictating
what sources a journalist will seek out for a story, and may deter a
journalist from seeking out voices that may draw too many questions about
the argument a reporter is trying to construct in a story.
Journalists also rely on official sources to frame a story – to put it into
context and give an event meaning (Durham, 1998). Because journalists are
pressed for time, they often opt for sources whose official status give
them verifiable credibility and immediate accessibility (Powers & Fico,
1994). Powers and Fico examine the factors that contribute to journalists'
selection of sources and find that a reporter's own orientations have the
most profound impact. They speculate that because their study looks at
larger newspapers, the reporters questioned had greater experience and
leeway in choosing the sources in their reporting. However, they also
theorized as journalists become more educated about their beat, their need
to rely on official sources decreases – perhaps driven by skepticism of
those same sources. Powers and Fico suggest additional research should be
conducted to explore source usage at smaller newspapers.
Role of community media
The community integration theory holds that newspapers have long played a
role in the creation and defining of a community, but that theory has
largely been abandoned in recent years (Stamm, Emig & Hesse, 1997). Stamm
and colleagues found that in modern society the role of local media in
defining a community may be overestimated, and that interpersonal
communication is the "primary mechanism for community integration of the
individual" (p. 105). So, if the local media is no longer considered to be
a key factor in the integration of a community, what role do the media play?
Community pluralism may have profound effects on a newspaper's willingness
to report on conflict (Olien, Donohue, & Tichenor, P.J., 1968, 1980, 1995).
Olien and colleagues argue that the media are not "the independent,
self-styled social agents that either they or members of the public may
imagine them to be" (p. 217, 1980). The journalist is but one strand in the
fabric of society, and journalists who try to remove themselves risk
unraveling the system.
The work of Donohue, Tichenor and Olien (1995) suggests the local media
sometimes serve as guard dogs, an analogy which suggests the media "perform
as a sentry not for the community as a whole, but for those particular
groups who have the power and influence to create and control their own
security systems" (p. 115). This dynamic, they assert is particularly
apparent in a homogenous community where those in power at the local level
are protected from anything viewed as an outside threat. Taken to an even
further extreme – when a small, homogenous community is dependent on
agriculture or one larger manufacturing firm – the media face great
pressure to operate as a kind of "newsletter" promoting the interests of
that entity. They hypothesize that when the media take on the role of guard
dog in a highly pluralistic society, they are less likely to be questioned
and challenged about their role.
If Donohue and colleagues' presentation of the guard dog theory is correct,
homogenous communities will be more likely to have newspapers operating
with the guard dog perspective. Whereas, community papers serving more
diverse areas – often in metropolitan settings – would be assumed to have
more pluralistic populations. To summarize, a paper with a "guard dog"
mentality might be predicted to use the voices of the official in a
community in their news reporting.
There are other factors that may determine the pressures the media can face
when operating in a community with clearly defined cultural meanings
(Berkowitz & TerKeurst, 1999). Newspapers in more pluralistic, larger
communities may be more inclined to report stories in a more adversarial
reporting style, and to include the voices of those challenging the
dominant power structure. Berkowitz and TerKeurst conclude that those in
the elite, especially in small, homogenous communities, hold power in
dictating the dynamics of a community, though there are many more nuances
that go into defining the cultural meaning of a geographic community. A
newspaper that deviates from the "preferred meanings of a geographic
community" (p. 130) is putting itself at economic risk.
Journalists may see their coverage as objective, unswayed by the powers of
the elite, when in reality, those in positions of power may have great
control on the information used in articles. In a study of campaign
coverage, Fico and Cote (1997) found reporters felt their coverage was
balanced, but a content analysis of their work revealed a tendency to give
incumbents greater opportunity to voice their positions than challengers.
Fico speculated that this seeming hypocrisy exists because journalists had
not defined how to be fair in their reporting.
Just as with Massey's civic journalism example, the stated ideals of the
community press appear to conflict with what scholars predict to be the
outcome when the media face a united pressure from the powerful and the
official in a community.
This study provides an ideal opportunity to test the guard dog perspective
in different weekly newspapers and see how it compares with the ideals as
outlined by community journalists to create a news product that is
representative of the voices of the ordinary people in a community. By
looking at papers of similar circulation size serving communities of
varying diversity, it will offer a chance to explore whether community
homogeneity plays a role in predicting the dominance of official sources in
dictating the content of community newspapers.
Hypotheses
This study will address the relationship between the diversity of a
community and the use of official and unofficial sources in framing the
news. This study will look solely at community newspapers, defined as
weekly newspapers with a circulation of less than 20,000. To test this,
three hypotheses will be used:
H1: The greater the diversity of a community served by a weekly newspaper,
the more unofficial sources are used in news stories.
H2: The greater the diversity of a community served by a weekly newspaper,
the greater the prominence of unofficial sources.
H3: The greater the diversity of a community served by a weekly newspaper,
the more space devoted to unofficial sources.
For the purposes of this research the diversity of a community served by a
weekly newspaper is theoretically defined as the number of non-white
residents of a community in the geographic circulation area defined by a
weekly newspaper. For H1, the use of official sources in news stories is
the use of sources in positions of authority and power in the news stories
contained in a weekly newspaper. Unofficial sources are the voices of those
not identified in the story by their title or position (Massey, 1998).
Therefore, it is predicted that weekly newspapers that are circulated in
areas with greater diversity will rely less on the use of official sources
in individual news stories, and have greater representation from unofficial
sources. Conversely, weekly newspapers serving communities with less
diversity are expected to use more official sources, and have a greater
reliance on those in power to interpret the news.
H2 makes predictions concerning the prominence of unofficial sources in
news stories, which is defined as the credibility attributed to sources of
unaffiliated citizens through where they are placed in news stories. The
logic is that where a source first appears in a story says something about
how important that source is deemed to be by the reporter (Massey, 1998).
It is predicted that weekly newspapers that are circulated in areas with
greater diversity will give greater prominence to the voices of
unaffiliated citizens in their news stories.
The final variable to be measured in this study is the amount of space
given to unofficial sources in news articles. For H3, it is predicted that
weekly newspapers that are circulated in areas with large populations will
give more space in news articles to the voices of unaffiliated citizens in
their news stories.
These theoretical linkages are based on a literature review of studies that
have examined the role of official sources in framing the news, the factors
that contribute to source selection from reporters and on theory explaining
the role that newspapers play in a community.
Method
To measure use of official and unofficial sources in weekly newspapers, a
content analysis of sources in 10 weekly newspapers in New York state was
conducted. Each paper has a circulation of less than 20,000. The editions
analyzed were all published in a time period beginning in September 2000
and ending in September 2001, for a total of 4 issues for each publication
(40 individual newspapers). The first issue was chosen at random from the
first month, and then a systematic sample was used to select three more
issues. The newspapers selected were based on a convenience sample of
weekly newspapers that are members of the New York Press Association. The
papers represent independent and corporate-owned papers serving communities
of varying levels of diversity. The NYPA annual directory was used to
measure the circulation figures for each paper. Coupled with this content
analysis, population figures obtained from the U.S. Census Bureau's 2000
Census were used to evaluate each paper based on the percentage of
non-white residents in the county where the newspaper publishes.
The independent variable for all three hypotheses in this study was the
same: the diversity of a community served by a community newspaper. This
variable is operationalized as the percentage of non-white residents in the
county, based on data collected in the 2000 Census. While Census figures of
smaller geographic units are available, county subdivisions – such as towns
and cities and Census subtracts – do not offer comparable figures nor do
they provide information on the diversity of the entire region in which a
newspaper operates.
All three hypotheses also used the sources contained in local news stories
as the recording unit, while the newspaper was the unit of analysis. A
local news story was operationalized as all non-advertising matter in a
news product. In a newspaper, this would usually include all staff-produced
news stories found in the first and "local" sections, but excluding
editorial pages, op ed pages, reader opinions, sports, routine business
data, society news and similar matter. It may include relevant features
produced by local staff reporters and syndicated and wire services (Riffe,
et al., 1998, p. 112). All printed material under a single headline or
subheading. Cutlines, charts and pictures also were included in the analysis.
The dependent variable for H1, the use of unofficial sources in news
stories, is operationalized as the number of sources categorized as
official sources versus unofficial sources contained in newspapers analyzed
by content analysis. The method of coding the use of official and
non-official sources is based on methods used by Massey (1998) in his
evaluation of the use of official and unofficial sources and by Voakes, et
al., (1996), which also measured source diversity. Sources were defined as
being official by attribution identifiers. As defined by Massey,
non-official sources were operationalized as "unknown public servants,"
"unknown criminal justice workers," "unknown educators," "non-governmental
and civic group workers," "private business workers," and "unaffiliated
citizens." Official sources were operationalized as "known public
officials," "known criminal justice officials," "known educators," "known
private-business officials," "professional and experts" and "affiliated
citizens." The operational definitions of these categories are further
explained in the coding protocol for this study. Each of these
subcategories was then collapsed into unofficial and official measures. The
number of official and unofficial sources was categorized for each article
and for each edition. Official and unofficial sources were then compared by
calculating the percentage of each type of source used by each
publication. The operational linkage for H1 predicts that a paper that
publishes in a community with less diversity will have a lower percentage
of unofficial sources.
For H2, the dependent variable of prominence is operationalized as the
paragraph number in which a unofficial source is first named in a news
article. Sources were coded based on the paragraph in which an attributed
statement from that source first appears – the logic being that the higher
in the article a source's attributed comments appear, the more newsworthy
that source's opinion was deemed by the reporter (Massey, 1998). The
operational linkage for H2 predicts that a paper publishing in a community
with less diversity would have a high value for the variable prominence –
meaning unofficial sources would be introduced later in the article than
official sources.
Space devoted to unofficial sources, the dependent variable in H3, is
operationalized as square inches of space attributed to official and
unofficial sources. This was calculated by measuring the full and partial
lines of text attributed to a source either through direct quotation or
paraphrase. Lines of copy were measured in square inches by measuring the
depth and width of the identified lines of attribution and multiplying
those numbers. For example, if three lines of attribution had a depth of .5
inches and a width of 2 inches, they would be multiplied and coded as 1
(square inch). All numbers are coded to the nearest inch, with .5 equaling
1 and 1.4 equaling 1. The operational linkage for H3 predicts that a paper
publishing in a community with less diversity would devote less space to
information attributed to unofficial sources than to official sources.
These variables were measured by coding each source for all three
variables, and then collapsing this information by newspaper to produce
these categories: population, percent official sources, average prominence
of official sources, average prominence of unofficial sources, average
space official and average space of unofficial. All variables were then
compared using Pearson's correlation coefficient.
A pretest of the coding protocol was conducted using two coders on all
three variables. The pretest of prominence yielded intercoder reliability
with a Scott's pi of 1. The pretest of source identification yielded a
Scott's pi value of .722. The variable of space also yielded a strong
intercoder reliability, with a Scott's pi of 1.
Two coders were used to analyze more than 10 percent of the data in this
study, and the intercoder reliability was acceptable for all variables. The
reliability of prominence was found to have the greatest agreement among
coders with a Scott's pi value of .88. The agreement on the variable of
space devoted to sources was also good, with a Scott's pi of .77. The
lowest agreement was in the coding of official and unofficial sources, with
a Scott's pi agreement of only .58.
Results
Content analysis yielded a total of 659 sources from the 40 newspapers
coded. The variation in sources ranged from a high of 193 sources for the
Bayside Times, to a low of 23 sources for the Journal and Republican. To
test the variables outlined by the three hypotheses the sources were then
collapsed by newspaper to measure five content-related variables with a
sample size of 10. A sixth variable of population based on 2000 Census data
was also included.
Table 1 about here
The variables used to compare the 10 newspapers in this study are shown in
Table 1. The range in diversity had a high of 55.9 percent for the
Bayside Times, to a low of 2.8 percent for the Citizen Outlet. The
variation in diversity between the communities these papers serve offers an
ideal opportunity to explore whether there is a relationship between
community diversity and the variables used to measure the use of official
and unofficial sources. The average percent of nonwhite residents of the
counties in which the newspapers were based was 16.7 percent, with a
standard deviation of 15.37.
Sources in this study were categorized as being either official or
unofficial. Overall, the percent of official sources averaged 63.8 (sd =
9.7), and shows that, on average, the newspapers studied chose to use more
official sources than unofficial sources. Only one newspaper, the Citizen
Outlet used fewer official sources – 44 percent – than unofficial sources.
The paper with the highest percent of official sources was the Staten
Island Register, which used official sources in 76 percent of the cases
evaluated.
The average of prominence of official sources had a mean of 5.6 (sd = 1.2),
which means that on average information attributed to official sources
appeared between the fifth and sixth paragraph in a news story. The
standard deviation shows there was little variation in prominence between
newspapers. The Patriot and Free Express used official sources the soonest,
averaging 3.6, and is in the second-most homogenous counties of the
newspapers evaluated in this study. Likewise, the Bellmore Herald
introduced attributed statements by official sources the latest, averaging
7.5, and had the third-highest population diversity studied.
The average prominence of unofficial sources, an average of 5.6 (sd = 1.6),
was almost equivalent to the use of official sources. The Journal and
Republican introduced unofficial sources the soonest of the papers
evaluated, with an average of 3.2, while the Bellmore Herald introduced
unofficial sources later in its articles, with an average of 8.3.
The measure of space devoted to official sources had an average of 4.9
square inches, and a standard deviation of .79, showing little variation
between the papers studied. The highest was again the Bellmore Herald at
5.8, while the lowest of was the Citizen Outlet with 3.5.
The average amount of space devoted to lines of unofficial sources had a
mean of 5.8 square inches, and a standard deviation of 2.1. The Journal and
Republican devoted the greatest amount of space to unofficial sources with
9.7 square inches, while the Patriot and Free Express gave only 3.3 square
inches of space to unofficial sources on average.
To test the three hypotheses in this study, Pearson's correlation
coefficients were conducted comparing the six variables used in this study.
Since all three hypotheses used community diversity as the independent
variable, it is the relationship between this variable and the others that
is of the most interest.
Table 2 about here
H1 was not supported. The data yielded no support for the relationship
between the percentage of official sources and community diversity. The
Pearson correlation coefficient was not statistically significant. As
previously stated, with one exception, all the newspapers used official
sources for a majority of the attributed facts in news articles.
H2, predicting a relationship between prominence of official sources, was
partially supported. The Pearson r between community diversity and the
prominence of official sources was .64 (p < .05) and suggests a strong,
positive relationship between the diversity of a community and where a
newspaper introduces attributed information from an official source. This
correlation shows that papers serving more-diverse communities had a
greater value for this variable than papers in less-diverse areas. This
relationship means that a homogenous-community newspaper is more likely to
introduce facts attributed to official sources earlier in a story than a
heterogenous-community newspaper.
In order for H2 to hold, however, it would follow that a diverse-community
newspaper would give lower prominence to unofficial sources and greater
prominence to unofficial sources when compared to newspapers serving
less-diverse communities. When diversity and the prominence of unofficial
sources were paired, however, a statistically significant relationship (p <
.05) was found with a Pearson's r of .69. However, this relationship does
not add support to H2, because it shows that a newspaper serving a more
heterogeneous community also appears to give lower prominence to unofficial
sources, when compared to newspapers serving more homogenous communities.
When the significant relationships found in these two pairings are
considered together, they demonstrate that a newspaper serving a more
diverse community gives less prominence to the attributed statements of
both official and non-official sources.
Another significant relationship was found between the variables not
related to the predictions made in this paper: A significant relationship
(p < .05) exists between the prominence of official sources and the
prominence of unofficial sources. A Pearson's correlation coefficient was
found to be statistically significant, with a Pearson's r of .76. The
relationship is explained in Table 1, which shows that a newspaper that
introduces attributed information high in an article will do so whether
that information is attributed to official or non-official sources. Taken a
step further, this relationship appears to be influenced by the diversity
of the community in which a newspaper exists, so that a newspaper in a more
homogenous community will introduce information attributed to a source
sooner in a story than a newspaper in a more-diverse community.
H3 was not supported. The correlation of the variables of diversity and
space devoted to unofficial sources was not statistically significant.
Discussion
If weekly newspapers share a common ideal, then it is in the desire to
report the voices of the everyday people in a community. These routinized
forces predict that any community newspaper would uniformly include the
voices of the unofficial people – the rank and file employees of
government, the school teachers and the owners of small "mom and pop"
businesses. However, another influence on media content, community
diversity, predicts that the use of unofficial sources will vary based on
the homogeneity of the community being served.
This study attempts to gauge how community newspapers use the voices of the
official and unofficial in their news stories, and to explore whether that
relationship changes based on the type of community in which the newspaper
publishes. While there has been much research looking at how larger, daily
newspapers use official and unofficial sources, little research could be
found that explores these relationships at small, weekly newspapers. The
predictions made in this study were based largely on theory proposed by
Donohue and colleagues (1995), which suggests that papers serving more
homogenous communities are more likely to act as "guard dogs" rather than
watchdogs. In short, papers in these communities would be more likely to
protect those in positions of power, and less likely to represent the
voices of those who might threaten the dominant power structure.
Three hypotheses were posed to evaluate how community newspapers used
official and unofficial sources in reporting the news. Community diversity
was the independent variable in all three instances. H1 predicted that a
paper publishing in a more-diverse population area would use more
unofficial sources in news stories. H2 posed that a newspapaer in a
more-diverse area would give greater prominence to information attributed
to the unofficial. Finally, H3 suggested a paper in a more-diverse area
would give more space to unofficial sources.
These predictions were tested by conducting a content analysis of 659
sources from 40 issues of 10 different weekly newspapers in New York state
that served communities of diverse populations. Four samples of each
newspaper were randomly selected using a systematic sample of issues
published over the course of one calendar year. The sources in each
newspaper edition were categorized as being official or unofficial, and
were also evaluated based on where information attributed to them appeared
in a news article and on the number of square inches devoted to their
comments. These sources were then evaluated by newspaper examining six
variables: percentage of non-white residents in the county in which the
newspaper publishes, percent of official sources, average prominence of
official sources, average prominence of unofficial sources, average space
devoted to official sources and average space devoted to unofficial sources.
The results of this study found no evidence to support H1 or H3. Overall,
community newspapers – with one exception – were found to favor the voices
of official sources in their news articles, which is consistent with
previous research. These findings did not appear to vary based on the
diversity of the community the newspaper serves, suggesting a uniformity in
media content perhaps governed by routinized forces (Shoekmaker, et al., 2001).
There appears to be support for the prediction of the second hypothesis,
that unofficial sources would receive greater prominence in papers serving
more-diverse communities. However, H2 is called into question because this
same treatment was given to official sources. A statistically significant
positive Pearson's correlation coefficient was found for both the
prominence of official and unofficial sources with diversity as an
independent variable. Heterogeneous-community newspapers were found to cite
sources lower in a story than homogenous-population newspapers, and this
occurred whether that source was official or unofficial.
This paper has brought to light a number of issues that are of importance
in explaining the role that the community press plays in the places it serves.
First, there appears to be a relationship between the number of sources in
a newspaper and the diversity of the community in which a newspaper
circulates. A newspaper serving one of the boroughs of New York, for
example, was found to use more than 193 sources over the course of four
editions, compared with as few as 23 in four editions of a paper that
served a more rural community. A post hoc analysis comparing the variables
of total number of sources with community diversity found a significant
relationship (p < .05) with a Pearon's r of .75. This difference may point
to the fact that there is a profound difference in the resources that
smaller, more rural community newspapers have when compared to their
big-city counterparts. It's important to note that most of the
homogenous-community newspapers in this study were independent papers in
which the publisher owned only one newspaper. In contrast, several of the
papers serving suburban and metropolitan areas were one of a group of
papers serving a contiguous area in which one story may run in more than
one newspaper in that group.
Second, the use of official and unofficial sources is only one way to
evaluate whether a community newspaper is serving as a "guard dog" or
whether it has the courage to challenge those in positions of power. The
type of content analysis conducted in this study did not allow evaluation
of the framing of stories or whether story topics are being written to
cater to the official or to serve the everyday readers in a community.
Third, more research should be done to compare the use of unofficial
sources between weekly newspapers and larger, daily newspapers. Community
journalists espouse a philosophy of wanting to include the voices of the
entire community, so it is crucial to explore whether this ideal is
manifest more significantly in their content.
Fourth, it is hoped that more research will be done to explain the
relationship found in this paper regarding where in a news story a
newspaper introduces attributed information. This relationship, and the
variability regarding the number of sources used in newspapers serving
varying communities may hint at a disparity between the resources that
exist at community newspapers serving rural and metropolitan areas. To
determine this, more research should be done to compare community
newspapers based on size, staffing and other factors that would help define
the prosperity of papers serving rural and metropolitan areas.
As noted, the community press is one of the fastest-growing segments of
this nation's newspaper industry, and yet there seems to be a dearth of
scholarly work that would serve this important segment of the print
newspaper industry. It is hoped this paper will inspire others to delve
into exploring this fascinating aspect of journalism.
References
Berkowitz, D., & TerKeurst, J. (1999). Community as interpretive community:
Rethinking the journalist-source relationship. Journal of Communication,
49, 125-136.
Donohue, G. A., Tichenor, P. J., & Olien, C. N. (1995). A guard dog
perspective on the role of media. Journal of Communication, 45, 115-132.
Durham, F. (1998). News frames as social narratives: TWA flight 800.
Journal of Communication, 48, 100-117.
Fico, F, & Cote, W. (1997). Fairness and balance in election reporting: The
1994 governor's race in Michigan. Newspaper Research Journal, 18, 50-63.
Hansen, K. A. (1991). Source diversity and newspaper enterprise journalism.
Journalism and Mass Communication Quarterly, 68, 474-482.
Lauterer, J. (2000). Community journalism: The personal approach. Ames, IA:
Iowa State University Press.
Massey, B.L. (1998). Civic journalism and unofficial sourcing: Making
routine newswork of community connectedness. Journalism and Mass
Communication Quarterly, 75, 394-407.
'99 Facts about Newspapers. (1999). Reston, VA: Newspaper Association of
America.
Olien, C. N., Donohue, G. A., & Tichenor, P.J. (1968). The community
editor's power and the reporting of conflict. Journalism Quarterly, 45,
243-253.
Powers, A., & Fico, F. (1994). Influences on use of sources at large U.S.
newspapers.
Newspaper Research Journal, 15, 87.
Riffe, D., Lacy, S., & Fico, F. (1998). Analyzing Media Messages: Using
Quantitative Content Analysis in Research. Mahwah, NJ: Lawrence Erlbaum
Associates, Inc.
Salomone, K. L., Greenberg, M.R., Sandman, P.M., & Sachsman, D.B. (1990). A
question of quality: How journalists and news sources evaluate coverage of
environmental risks. Journal of Communication, 40, 117-130.
Shoemaker, P.J., & Mayfield, E.K. (1987) Building a theory of news content:
A synthesis of current approaches. Journalism Monographs, 103, 1-36.
Shoemaker, P.J., & Reese, S.D. (1996). Mediating the Message: Theories of
Influences on Mass Media Content. White Plains, NY: Longman.
Shoemaker, P. J., Eichholz, M., Kim, E., & Wrigley, B. (2001). Individual
and routine forces in gatekeeping. Journalism and Mass Communication
Quarterly. 78, 233-246.
Stamm, K.R., Emig, A.G., & Hesse, M.B. (1997). The contribution of local
media to community involvement. Journalism and Mass Communication
Quarterly, 74, 97-107.
Stocking, S., LaMarca, N. (1990). How journalists describe their stories:
Hypothesis and Assumptions in Newsmaking. Journalism Quarterly, 67, 295-301.
Tichenor, P.J., Donohue, G. A., & Olien, C. N. (1980). Community conflict
and the press. Beverly Hills, CA: Sage.
Tuchman, G. (1978). Making News: A study in the construction of reality.
New York: The Free Press: A Division of Macmillan Publishing Co., Inc.
Voakes, P., Kapfer, J., Kurpius, D., & Chern, D. (1996). Diversity in the
news: A conceptual and methodological framework. Journalism and Mass
Communication Quarterly, 73, 582-593.
Tables
Table 1. Official source variables by weekly newspapers.
Newspaper Diversity* % official** Avg. prom.*** Avg. prom. Avg. space
**** Avg. space
official unofficial official unofficial
Bayside*****
Times 55.9 64 6.6 8.0 4.2 3.5
Staten Island
Register 22.4 76 6 6.1 4.6 6.3
Bellmore
Herald 20.7 55 7.5 8.3 5.8 5.4
Amherst
Bee 17.8 64 6.4 5.9 5.5 8.9
Messenger 15.2 70.5 6.3 5.4 4.2 5.1
Journal and
Republican 11.3 65 5.3 3.2 5 9.7
Mountain
Eagle 9.2 74 5.5 6.4 5.4 5.6
Country
Courier 8.2 69 4.5 4.1 4.9 4.9
Patriot and
Free Express 3 57 3.6 4.3 6 3.3
Citizen
Outlet 2.8 44 4.0 4.7 3.5 5.2
* Diversity of county in which newspaper circulates is measured as
percent of non-white residents, based on 2000 Census data.
** Sources were coded: 1 = unofficial source, 2 = official source.
*** Prominence was coded as first paragraph in which an attributed comment
from a source appears. A source with a high value for prominence would
appear lower in a news story than a source with a low value for prominence.
**** Lines for source was coded as the number of square inches of
paragraphs attributed to a source.
***** All papers are weekly newspapers with a circulation of less than
20,000. Four issues of each paper were selected using a systematic sample
of papers published between September, 2000, and September, 2001.
Table 2. Pearson correlation coefficients for unofficial source variables.
Variables 2. 3. 4. 5. 6.
Community .20 .64a .69 b -.23 -.18
1. Diversity* (.57) (.04) (.03) (.52) (.61)
.30 .04 .13 .17
2. Percent official** __ (.40) (.92) (.71) (.63)
3. Average .76 c .08 (.23)
prominence official*** __ (.01) (.83) (.53)
4. Average .06 -.32
prominence unofficial __ (.88) (.37)
5. Average .12
space official**** __ (.74)
6. Average
space unofficial __
* Diversity of county in which newspaper circulates is measured as
percent of non-white residents, based on 2000 Census data.
** Sources were coded: 1 = unofficial source, 2 = official source.
*** Prominence was coded as first paragraph in which an attributed comment
from a source appears. A source with a high value for prominence would
appear lower in a news story than a source with a low value for prominence.
**** Lines for source was coded as the number of square inches of
paragraphs attributed to a source.
a p < .05
b p < .001
c p < .05
Appendix A
Coding sheet
Variable
Variable name
coder
v1
Source identification
v2
Page number
v3
Story day (month and day)
v4
Newspaper name
Bayside Times = 1
Amherst Bee = 2
The Mountain Eagle = 3
Citizen Outlet = 4
Journal and Republican = 5
Messenger = 6
Country Courier = 7
Staten Island Register = 8
Patriot and Free Press = 9
Bellmore Herald = 10
v5
Source prominence ( 1,2,3…)
v6
Source identification
Unofficial source (1)
unknown public servants
nongovernmental
unknown business
unaffiliated citizens
Official sources (2)
government
law enforcement/criminal justice
education
known business officials
affiliated citizens
professionals and experts
v7
Space for source (Square inches text)
Appendix B: Coding protocol for official and unofficial sources
Introduction
This coding protocol is aimed at assessing the use of official and
unofficial sources in community newspapers. The analysis will evaluate
sources in news stories in 10 different community newspapers. Sources will
be coded based on their appearance in the story, their designation as
official or unofficial sources and the amount of space attributed to each
source. The following definitions are important in selecting and analyzing
the content under study.
News story
News stories are defined as all non-advertising matter in a news product.
In a newspaper, this would usually include all staff-produced news stories
found in the first and "local" sections, but excluding editorial pages, op
ed pages, reader opinions, sports, routine business data, society news and
similar matter. It may include relevant features produced by local staff
reporters (Riffe, et al., 1998, p. 112). All printed material under a
single headline or subheading. Cutlines, charts and pictures also should be
included in the analysis.
Source
A source is a person, or organization, who gives information to news
reporters. Sources are explicitly identified as such when news reporters
quote or paraphrase information from them in stories. The means by which
reporters publicly credit a source for story information is called
attribution. Such attribution is signaled when a person or organization's
name is linked in a story sentence with verbs denoting a person speaking,
such as "said," "claimed," and so forth. Attribution also may be made by
verbs denoting a source's state of mind, such as "thinks," "feels," "want,"
etcetera. Story information not clearly attributed to a source is assumed
to originate from a reporter's direct observations of actions and events.
Procedure
The following steps should be taken in the content analysis coding
described below (v stands for variable): (a) All relevant news stories are
evaluated based on the following variables; (b) all stories are marked by
one coder; (c) each source in a story is analyzed for specific
characteristics described below.
v1 Source identification. Arbitrarily assigned number based on order of coding.
v2 Page number.
v3 Story day (month and day)
v4 Newspaper code (1-10)
Bayside Times = 1
Amherst Bee = 2
The Mountain Eagle = 3
Citizen Outlet = 4
Journal and Republican = 5
Messenger = 6
Country Courier = 7
Staten Island Register = 8
Patriot and Free Press = 9
Bellmore Herald = 10
v5 Source prominence.
The source is coded (1,2…) based on the paragraph number in which a
statement or quote is first attributed to that source. A source, for
example, whose attribution is listed in the second paragraph would be coded
"2."
v6 Source identification
Sources are defined as named persons who appear in a news story and to
which facts are attributed either through quotes or paraphrase. Sources
will be coded as belonging to one of the following 2 categories, official
and unofficial. The following definitions as defined by Massey (1998) are
to be used in deciding into which category a source is classified.
Unofficial sources (1)
Includes the following categories:
Unknown Public Servants
Rank-and-file employees of any governmental department, division, agency,
program or commission; police patrolmen and women; sheriff's deputies;
public-school teachers. University/college professors not speaking on
their fields of interest or quoted as subject-matter experts.
Nongovernmental
Presidents, administrators, chiefs, heads, directors, board members, etc.,
of any nongovernmental or nonprofit agency, organization or institution
(i.e., United Way, Girls and Boys Club, grassroots and/or community-based
development agencies); festival organizers or organizing groups. Employees
or spokespersons of any of the above.
Unknown Business
Owners, managers of lesser known businesses or companies (i.e., locksmith,
beautician, mom-and-pop shop); rank-and-file employees of lesser known
businesses or companies.
Unaffiliated Citizens
Parents; residents; accident, disaster and crime victims and their friends
and relatives, or witnesses to accidents, disasters or crimes; lesser known
criminal suspects and convicts; any public-school, university or college
student; people as interns; volunteers; club members or officials; members
of informal/unnamed resident/neighborhood groups. Also, any person for
whom a news story does not explicitly identify by institutional,
organizational or governmental affiliation.
Official sources (2)
Includes the following categories:
Government
U.S. senator or representative; state senator or representative; governor
or cabinet member; county commissioner or county administrator; city
commissioner; mayor or town administrator. Also, any governmental
department, division, agency or program supervisor, director chairperson,
etc. – and their immediate assistants and deputies.
Law Enforcement/Criminal Justice
State attorneys or prosecutors; public defenders; judges; police chiefs;
sheriffs; investigators and ranked officers; notorious or widely publicized
suspects and convicts.
Education
School-district superintendent; school board members and chairperson;
public-school principals; university or college presidents; members or
chairpersons of university/college governing boards; State University
System regents (board president or chairperson).
Any candidate for any elective governmental office, or for
university/college president.
Any spokesperson for any of the above.
Known Business Officials
Owners or managers of prominent or widely known businesses, corporate and
private-sector utilities (i.e., cable TV company, phone company, large or
nationally/chain owned grocery or department store).
Any spokesperson for any of the above.
Affiliated Citizens
Activists for any environmental, political, religious or social cause;
lobbyists; trade union leaders, etc.; members and officials of formal/named
resident/neighborhood groups.
Professionals and Experts
I.e., lawyers, medical doctors, developers; university/college professors
speaking on their fields of interest; analysts; researchers; pollsters.
v7 Space for source (lines of text)
The lines of text attributed to each source are to be measured. The lines
are defined as the number of full lines of text attributed to a source
either through direct quotation or paraphrase. Lines of copy are measured
in square inches by measuring the depth and width of the identified lines
of attribution and multiplying those numbers. For example, if three lines
of attribution had a depth of .5 inches and a width of 2 inches would be
multiplied and coded as 1 (square inch). All numbers are coded to the
nearest inch, with .5 equaling 1 and 1.4 equaling 1.
|
