Corporate Newspaper Structure and Editorial Page Vigor
By David Pearce Demers
Assistant Professor
Department of Journalism
University of Wisconsin - River Falls
River Falls, Wisconsin 54022
715/425-3169 (office) / 612/490-5829 (home)
E-mail: [log in to unmask]
Demers is assistant professor of journalism at the University of Wisconsin
in River Falls. This research was supported in part by Grant
#1450-5-94
from the UW-RF Institutional Studies, Research and Grants Committee.
Abstract
Many critics contend that corporate newspapers are less vigorous
editorially than entrepreneurial newspapers because they are more concerned
about the bottom line than information diversity. This study, which
involves a national probability survey of daily newspapers, fails to
support that belief. Corporate newspapers publish a larger number of
local
editorials and letters to the editor, and a larger number and proportion
of editorials and letters that are critical of mainstream groups and
institutions. From a broader perspective, these findings may be
interpreted as supporting theories which hold that the pace of social
change quickens as social systems become more structurally pluralistic.
Number of words: 98
Corporate Newspaper Structure and Editorial Page Vigor
Many mass communication scholars and professionals believe the growth of
the corporate form of organization in the U.S. daily newspaper
industry
during the 20th century1 has reduced diversity in the marketplace of
ideas.
2 Critics have charged, among other things, that corporate newspapers
publish fewer editorials about local issues, publish fewer editorials
that
are critical of mainstream groups and ideas, and exercise greater
control
over their editorial editors and the editorials they write. Corporate
newspapers are less vigorous editorially, the critics contend, because
they
are afraid of offending advertisers, sources or readers, who may pull
their advertising, buy fewer newspapers, or complain about content.
Despite the popularity of the critical model, only three empirical
studies that have examined the effects of ownership structure on
editorial-page content support it. Most studies have found few differences
between chain and independent newspapers D the most frequently used
empirical measure of organizational complexity D and a fair number even
suggest that chain newspapers are more, not less, vigorous editorially
(see
review below).3 Additional research is necessary to clarify the effects.
But even if that research shows that chain or corporate newspapers are
less vigorous editorially, the critical model would still suffer from
an
important theoretical deficiency: It fails to account for social
change.
Taken to its logical extreme, the critical model implies that the
content
of corporate newspapers inhibits or impedes social change, a
supposition
that runs contrary to other empirical research which shows that mass
media
often play an important role in promoting social movements that
challenge
dominant authorities or ideas (e.g., women's and civil rights
movements).4
Is the editorial-page content of the corporate newspaper less critical of
mainstream groups and ideas than the content of entrepreneurial
newspapers,
as the critics contend? Or can corporate newspapers under some
circumstances publish news that is more critical of established groups and
that promotes social change, as some of the empirical research has
shown?
The purpose of this paper is to answer these questions by empirically
testing a model that attempts to explain the effect of corporate
newspaper
structure on editorial-page diversity. For purposes here, corporate
newspaper structure is conceptually defined as an organization that has
(1)
a clear-cut division of labor, (2) a hierarchy of authority, (3) rules and
regulations, (4) formalistic impersonality, (5) employment based on
technical qualifications, (6) rationality, or a high degree of efficiency,
and (7) a complex ownership structure (e.g., chain ownership, public
corporation).5 For heuristic purposes, the corporate newspaper may be
contrasted with the entrepreneurial newspaper, an ideal type that is
structurally simple and is owned and managed by the same individual or
family. But, ideally, corporate newspaper structure should be measured
as
a continuous variable.
Briefly stated, the theoretical model tested in this study holds that
corporate newspapers, like all mainstream media, are institutions that
play
an important role in maintaining the broader social structure (i.e., they
are social control agents). However, corporate newspapers are
expected to
be more, not less, critical of the established power structure than
entrepreneurial newspapers. Corporate newspapers are expected to publish
more staff-written editorials, more local editorials and more
editorials t
hat are critical of local mainstream groups and elites, both in terms
of
volume and proportion of total content. Corporate newspapers are
expected
to be more critical because increasing role specialization and
professionalism within the organization insulates editorial editors
from
political pressures, both outside and inside the organization. Data
from a
national probability survey of journalists at daily newspapers will be
used to test the hypotheses.
Literature Review
A review of the literature produced 16 studies that have examined the
effect of organizational structure on editorial-page content or staff.
All
employed chain ownership (chain v. independent newspapers) or some variant
of it (e.g., number of newspapers in chain) as the independent variable.[1]
No studies were found that use other measures of organizational
structure.
Of the 16 studies examined, three generally support the critical model,
seven show no relationship or have mixed findings, and six suggest
that
chain organizations are more vigorous or create conditions conducive to
greater diversity.
The earliest study was published in 1956 by Borstel, who wanted to know
whether "home-owned, non-chain papers show a greater interest in local
affairs of public interest than chain papers where the owners live
hundreds
or thousands of miles away," or whether "chain papers, because of their
greater financial strength, show greater forthrightness, greater
tendencies
to speak frankly, regardless of consequence, on local questions."7 He
content analyzed editorials, columns, letters to the editor and
cartoons
over a six-week period in 20 small dailies located in northern cities
with
under 25,000 population. He found "no consistent differences" by
ownership
structure.
In 1971, Grotta published the findings from a study which examined the
impact of ownership on size of editorial staff, size of news hole,
percentage of local news, size of the editorial-page news hole, and the
percentage of editorials as content. Regression analysis found no
significant differences between independent and chain-owned newspapers.8
Four years later another study reached similar conclusions. Wagenberg
and
Soderlund studied Canadian newspapers and found no correlation between
ownership structure and slant in the treatment of competing political
parties or the number of articles written about a variety of editorial
themes, including welfare, federalism, and tax reform.9
In contrast to those studies, Wackman, Gillmor, Gaziano and Dennis
examined newspapers' editorial endorsements of presidential candidates from
1960 to 1972 and concluded that chain newspapers exhibit a high degree of
homogeneity.10 "Clearly these data run counter to the insistence of
chain
spokesmen that their endorsement policies are independent from chain
direction. At an overt level, in terms of formal structural controls,
this
may be true, but at an informal level questions should be raised about the
degree to which hiring practices, management procedures and peer pressure
push chain newspapers toward uniformity of editorial posture."11 In
1977,
a quasi-experimental study by Thrift also disputed the comments of the
chain spokespersons.12 He compared editorials in 24 West Coast chain
and
independent newspapers before and after the chain-owned newspapers
were
purchased by a chain. He found that after the purchase, newspapers
that
became part of a chain were less likely to write editorials that dealt
with
topics of controversy, local or otherwise, and were less likely to write
"argumentative" (as opposed to explanatory) editorials. In contrast,
independently owned newspapers posted significant increases on these
measures. He concluded that "independently owned daily newspapers'
editorials do become less vigorous after the newspapers have been purchased
by chains."13
In contrast to Thrift's findings, a 1979 survey by the American Society of
Newspaper Editors found that editors at chain-owned newspapers were more,
not less, likely than those at independently owned newspapers to: (1)
take
stands that would be opposed by their publishers, (2) choose who their
newspaper would endorse in a national election and (3) say they never
had
to check with a newspaper's headquarters or owner before taking a
stand on
a controversial issues.14 A study by Goodman study three years later
also
produced results that generally supported the ASNE study.15 He read
three
months of editorials in 45 chain-owned and 25 independent Illinois
weeklies
and found that, although independent newspapers published a slightly
higher proportion of editorials about local and state subjects,
independent
papers published fewer editorials overall. The chain weeklies actually
published more editorials, more column inches of editorials and made
more
political endorsements than independent weeklies. Daugherty's Ph.D.
dissertation in 1983 also supported these studies.16 Examining the
editorial content of 36 chain and 32 independent daily newspapers, he
reported that chain papers published more letters to the editor and more
editorials, and had more editorials about local issues. No
differences
were found in the number of presidential endorsements.
A book edited by Ghiglione in 1984 that presented the findings of 10
different case studies of newspapers purchased by chains drew mixed
results
about the effects of ownership structure.17 The authors concluded that in
three cases the newspapers had improved, in three cases there were no
significant changes, and in four cases they deteriorated. In 1986, St.
Dizier reported the findings from a survey of editorial page editors,
which
found that chain newspapers were more likely to endorse the Republican
candidate for U.S. president in the 1980 election (i.e., Reagan).18
St.
Dizier also found that chains were more likely to have Republican
publishers. However, in the same year a study of 51 California newspapers
by Rystrom found that chains were more likely to endorse Democratic
candidates and that the gap had widened from 1970 to 1980.19
In 1988, Hale studied the editorial-page content of 28 daily newspapers
when they were independently owned and after they had been purchased
by a
large chain (groups that own six or more dailies).20 He concluded
that for
most of the papers the change in ownership resulted in "only modest change
and slight improvement or deterioration."21 The papers published about
the same number of editorial pages, editorials, and letters to the
editor
after the conversion. For only one of 16 measures was there a
significant
change: Chain newspapers published slightly fewer miscellaneous
articles.
In 1988, Romanow and Soderlund also reported that the purchase of The
(Toronto) Globe and Mail D which is considered to be Canada's "national
newspaper" D by the often-criticized Thomson Newspaper chain resulted
in
few editorial changes. The chain actually doubled the number of local
reporters after acquisition, and, editorially, it was somewhat more
vigorous editorially on international issues that involved the Britain or
the United States.22
In 1989 Gaziano expanded the Wackman et al. dataset to include later
elections and concluded that chains still tend to be homogeneous in their
presidential choices; however, as chains increase in size the degree
of
homogeneity declines.23 Busterna and Hanson, on the other hand,
contend
that there is little evidence to substantiate the chain homogeneity
argument.24 Akhavan-Majid, Rife and Gopinath found a high level of
agreement among editorial positions taken by Gannett newspapers compared
with a matched sample of independently owned newspapers; however, the
Gannett newspapers were far more likely to editorialize on the three
national issues studied and to oppose the positions taken by dominant
elites (e.g., President, Supreme Court).25 A more recent study by
Akhavan-Majid and Boudreau found that editors at chain newspapers are more
likely than their counterparts at independently owned newspapers to
say
that the role of their newspaper is to provide critical evaluation of
local
government performance and to function as a watchdog of business on behalf
of consumers.26 Large newspapers also were more likely to say the role of
their newspaper is to function as a watchdog of business. And, finally,
Wilhoit and Drew found that editorial editors at group-owned
newspapers are
far more likely than those at family or independently owned newspapers to
say publishers have no influence or very little influence when it
comes to
"determining the priority given to editorial topics."27
In sum, Table 1 shows that the weight of the empirical evidence fails to
support the critical model D that chain newspapers actually are
slightly
more vigorous editorially or have the capacity to be more vigorous
than
independent newspapers.
-------------------------------
Insert Table 1 about here
--------------------------------
A Structural Theory of Editorial-Page Content
The empirical literature on chain ownership and editorial pages is
illustrative. Caution should be used, however, in generalizing the
findings from these studies, for at least two reasons.
The first is that these studies employ only one indicator of
organizational structure D chain ownership. Although ownership is an
important indicator of organizational complexity, it is a very crude
measure,[2] and this shortcoming may contribute to the mixed findings in
the
literature.29 More appropriately, the growth of the chain newspaper
should
be conceptualized as a consequence of increasing social complexity and
economic competition. As social systems grow and become more complex
(i.e., become more structurally pluralistic), competition between mass
media for limited resources (e.g., advertising and audiences) increases,
which intensifies social and technological innovations that promote
via
economies of scale the growth of large-scale media organizations.30
Chain
ownership is one indicator of this process of organizational growth
and
chain, but only one. Other measures include division of labor and role
specialization, hierarchy of authority, rationality in decision-making,
formalized rules and procedures, and a highly skilled and educated
staff.[3]
The second problem with the empirical literature is that many of the
studies are based on faulty or unjustifiable assumptions. The argument
that corporate newspapers are less vigorous editorially, for example,
is
usually based upon the assumption that they place greater emphasis on
profits and less on product quality or the information needs of the
community. While there is some evidence to suggest that chains place more
emphasis on profits,[4] recent national probability surveys have found
that
corporate newspapers actually place less emphasis on profits as an
organizational goal and more emphasis on product quality than
entrepreneurial newspapers.33 These studies and others also suggest
that
the assumption of an inverse relationship between profits and product
quality (i.e., a zero-sum formula) is conceptually and empirically
flawed.
Research suggests, in fact, that highly profitable companies appear to
spend a larger proportion of their profits on editorial production34
and
that corporate organizations have a greater capacity to pursue more
goals.
35
To be sure, some newspapers that exhibit the characteristics of the
corporate form of organization do conform to critical model, and case
studies can be produced to document such cases.36 But these cases are
the
exception rather than the rule. Theoretically, the transition from
the
entrepreneurial to the corporate form of organization under most
conditions
would be expected to produce a more, not less, vigorous press. Two major
reasons may be cited to support this proposition.
The first is that corporate newspapers themselves are more likely to be
located in communities that contain more social conflict and criticism
of
dominant groups and values systems.[5] A well-documented research
finding is
that newspapers in homogenous communities contain less conflict and
criticism; in fact, weekly newspapers and small dailies are often seen as
"boosters" or "cheerleaders" for the local community.38 The amount of
social conflict and criticism of mainstream institutions and values in
these community newspapers is low in part because the community contains
a
limited number of alternative or challenging groups and
organizations.[6]
Small communities also do not encourage or tolerate a wide range of
behaviors, opinions or values, at least openly. Elites in small,
homogenous systems share similar interests, values, goals and world views.
Decision-making relies more heavily on consensus than debate. But even
when conflict emerges between different groups of elites or
challenging
groups, media in small communities tend to limit reporting of such
conflicts.40 Social conflict usually is perceived to be disruptive of
community solidarity, and these communities are not structurally
equipped
to deal effectively with open conflict. Conflicts often are handled
informally and decisions on crucial issues are reported by local media
after the fact.41
In contrast, social conflict is a much more common feature of large,
pluralistic communities because they contain a much greater variety of
special interest groups competing for limited social, political and
economic resources.42 Decision-making in such communities is expected to
take into account diverse perspectives and views, and such communities
are
structurally organized to deal with conflict, having mechanisms such
as
boards of inquiry (e.g., racial discrimination commissions, civilian
police
review boards), formal labor-management negotiators, formalized grievance
procedures, and administrative law judges. Although stories and
editorials
that contain conflict or criticism are often viewed as threatening to the
social order, such stories often play a significant role in
contributing to
system stability because they introduce alternative ideas or innovations
that enable organizations and institutions to adapt to changing
conditions.
As Donohue, Tichenor and Olien point out:
Conflict control may include the generation of conflict situations as well
as the direct dissipation of tension. This principle is widely
recognized
in the political realm ... . Media reporting of a clash between
scientific
opinion on supersonic transports and governmental policies regarding such
technology represents a generating of conflict. From a systems
perspective, such reporting is functional for maintenance of the total
system... .43
The second reason corporate newspapers would be expected to generate
editorial content that is more critical of dominant institutions and
values
is that their publishers and editorial staffs are more insulated from
special interests and political pressures. The owners and top managers
at
corporate newspapers are more insulated because (1) they are less
likely to
grow up in the community their newspaper serves, (2) they work at the
newspaper for a shorter period of time, (3) they are oriented to the
larger
corporation, not the local community, and (4) the decisions they make are
more heavily influenced by professional norms and values, which place
a
higher premium on truth and criticism than on local parochial
interests.
Publishers and managers who have spent a lot of time in the community
their newspaper serves often develop close personal ties to local elites
and organizations. Such ties may foster a greater concern with the
issues
and problems facing the community; however, they also may lead to
greater
constraints on the editorial process. Friendships create obligations,
and
the ability of the newspaper to impartially report on controversial
issues
or matters, especially those that portray local social actors in a
negative
way, may be compromised to the extent that a top-level manager is highly
integrated into a community. The effect of such ties is particularly
acute
in a small town, because the local entrepreneurial newspaper depends on a
smaller number of advertisers for its livelihood. While it is true
that no
newspaper can afford to alienate all or a substantial number of its
advertisers and expect to be financially self-supporting over a long
period
of time, corporate newspapers are more financially stable,44 which means
they are less dependent upon any single advertiser.
Top-level managers of corporate organizations also are more insulated from
political pressures because they spend less time working at those
newspapers and move more frequently from job to job.45 Many of these
managers, including top-level editors, are interested in climbing the
corporate ladder. This means they must be oriented to the larger
corporate
or chain organization, not the local community.46 An orientation to the
corporation may lead to less concern with local issues, but it also
reduces
the probability that local special interest groups will unduly influence
the news production process, giving corporate newspapers a greater
capacity
to criticize local groups.
Perhaps more important than living in a community and climbing the
corporate ladder, though, is the effect of professional norms and values on
the editorial process. Professional codes of ethics define public
conflict and criticism as newsworthy and condemn news that promotes local
parochial interests over truth and the public good.[7] While
professionalism
exists to some degree at most newspapers, at corporate newspapers
professionalism it is more advanced.[8] Contrary to popular belief, the
growth of corporate or bureaucratic institutions generally promotes D it
does not retard D the development of professional norms and values.49
The
division of labor and role specialization that accompanies the growth
of
large- scale organization facilitates the development of professional
norms
and values in part because those organizations have a larger number of
editorial employees who are structurally separated from workers in
other
functional areas. This separation facilitates the development of
specialized skills and knowledge as well as professional codes of conduct.
Large, complex, corporate newspapers, in fact, are much more likely to
have written ethical codes of conduct and to enforce them.[9]
Professional codes of conduct are designed, of course, to control the
behavior of professionals and, admittedly, they can, under certain
circumstances, inhibit the diversity of ideas.51 But they also help to
expand diversity by insulating journalists from special interest groups
(e.g., advertisers, politicians, government), who seek to use the media
to
serve exclusively their own needs and interests.[10] Professional
norms limit
or constrain the authority of noneditorial personnel in the news
production
process. When it comes to producing the news, the editors are the
experts, and only they have the authority to make decisions that affect the
editorial production process. Professional norms also help justify and
legitimate the role of journalists in producing news for the broader
society.
It is important to point out that the theoretical perspective presented
here is partially at odds with the view that professionalism leads to
homogeneity and standardization, not diversity. For example, Glasser
argues that professionalism
means quite the opposite of diversity. Whereas the goal of diversity is to
foster an appreciation for differences in experience and therefore
differences in knowledge, the goal of a professional education is D in
effect and usually by design D to unify knowledge by glossing over
differences in experience. Professionalism implies standardization and
homogeneity; it accounts not for differences among journalists but
for what
journalists have in common.[11]
Few scholars would disagree with the argument that professionalism can
contribute to standardization and homogenization D codes of ethics, for
example, are rules of conduct. However, this internal mechanism of
social
control (i.e., within the profession of journalism itself) is only
half of
the story. As noted above, professional norms and values also help to
insulate journalists from external controls (i.e., from noneditorial
news
workers and community groups and individuals). In fact, professional
codes
of ethics and norms account in large part for the ability of journalists
to collect legitimately news that occasionally criticizes those in
power
and the dominant ideologies.
In contrast to the critical model, the key strength of the structural
model presented here is that it helps to account for social change and
the
role that mass media often play in promoting such change. As social
systems (communities and the nation as a whole) become more pluralistic,
news media become more critical of traditional ways and established
institutions. Media reflect to some degree the diversity of the
communities they serve, and increasing role specialization and
professionalization, by-products of community growth, insulates
journalists
from outside pressures. The increased level of criticism that emerges
from these structural forces contributes to discourse that places
pressure
on existing institutions to change.
Although corporate newspapers increase the probability that editorial
content will be critical of established or mainstream groups, it is
important to point out that the structural or cultural changes that
sometimes result from these pressures are rarely radical. Social change
comes slowly. Research shows, in fact, that mass media are highly
responsive to political and economic centers of power and promote values
generally consistent with capitalist ideals and elite interests.[12]
The
claims of challenging groups are circumscribed to a large extent by the
opposing interests of those groups in power. The gains of the civil
rights
and women's movements of the 1960s, for example, have not eliminated
economic, political and social disparities between the races and sexes.
Nevertheless, it would be a distortion of history to claim that these
movements have been ineffectual or that the mass media played no role in
promoting their goals. Real structural change has occurred within the
system during the 20th century (e.g., affirmative action standards,
anti-discrimination laws), and media have played an important role in
promoting and, later, legitimizing such changes.
Hypotheses
The theoretical model outlined above leads to the expectation that, The
more a newspaper exhibits the characteristics of the corporate form of
organization: (H1) the greater the number of editorials and letters to
the
editor; (H2) the greater the frequency and proportion of editorials
written by local staff; (H3) the greater the frequency and proportion of
editorials and letters to the editor about local issues; (H4) the
greater
the frequency and proportion of editorials and letters to the editor
that
will be critical of mainstream sources or institutions.
To reiterate, the primary logic behind the first, second and third
hypotheses is that corporate newspapers D because they are larger, more
structurally complex and benefit from economies of scale D have more
staff
and resources to generate staff-written editorials on local topics and
more
space to print editorials and letters to the editor. The assumption here
is that organizational structure is a more important determinant of
the
volume and proportion of local editorial content than social or social
psychological ties to the local community stemming from long-term
residence
or community involvement. The fourth hypothesis is based on two
structural arguments: (1) Corporate newspapers are more likely to be
located in pluralistic communities, which contain more social conflict
and
criticism of dominant groups and values systems, and (2) Corporate
newspapers are more insulated from external political pressures because
their owners and managers are less likely to grow up in the community
their
newspaper serves, are more likely to work at the newspaper for a shorter
period of time, are oriented to the larger corporation as opposed to
the
local community, and are more influenced by professional norms and
values,
which value truth and criticism over local parochial interests.
Method and Measures
In September 1993, a two-page, 66-item questionnaire was mailed to the
highest ranking manager (e.g., publisher, general manager), the highest
ranking editor and a police reporter at 250 daily newspapers randomly
selected from the 1993 Editor & Publisher International Yearbook. The
mailings to the highest ranking manager and top editor were personally
addressed; the mailing to the reporters was simply addressed "police
reporter" because no list of names exists. Police reporters were included
in the sample to obtain a sampling of positions in the lower levels of
the
organization's hierarchy. They, rather than another reporting
position,
were selected because virtually every daily general circulation
newspaper
has a police reporter or someone who performs that function.55
Two follow-up mailings were conducted.56 Of the 750 questionnaires
mailed, responses were obtained from 409 journalists at 223 newspapers,
for
a total response rate of 55 percent. There were no significant
differences in response rates for the three groups (top manager, 52%; top
editor, 56%, and police reporter, 55%).57 There also was no
significant
correlation between the type of respondent (top manager, top editor,
reporter) and newspaper circulation (r=.01).58
In June 1994, another mailing was made to the 223 newspapers who responded
to the mail survey discussed above, asking them to send tear sheets of the
editorial and op-ed pages for the most recent Wednesday and Thursday
editions of the newspaper (self-addressed, postage-paid envelopes were
enclosed). Two consecutive days were sampled to reduce the burden of
complying with the request and, hopefully, to increase the response rate.
Two weekdays were selected because almost all newspapers that publish
on
weekends publish editorial pages in those editions but not weekday
editions
(i.e., sampling weekend dates may have led to an overestimate of the
quantity of editorials in smaller newspapers). Newspapers that did not
respond to the initial tearsheet mailing were then contacted by
telephone.[13]
Altogether, 198 newspapers, or 87 percent of the 223 newspapers in the
sample, responded. The issues sent covered the period from early June
to
mid-August.
Although individuals responded to the mail questionnaire, it is important
to point out that the newspaper D not the journalist D is the unit of
analysis. To conduct such an analysis, the findings were aggregated for
each newspaper that had more than one respondent. For continuous
measures
(i.e., ordinal, interval and ratio level measures) and dichotomous
nominal
measures,60 the final value used in the analysis represented the mean
of
the ratings given. In cases where the values for one of the
respondents
was missing (e.g., failure to answer a question), the values of the
other
respondent(s) were substituted. No nominal variables containing more
than
three values were included in this analysis.
Using Weber's conceptual framework as a guide,61 a corporate newspaper
index composed of five dimensions and 12 individual measures was
created.
The index and dimensions were derived from a factor analysis of 14
items in
a previous study.62 The first dimension or category is structural
complexity, composed of four individual measures (see Table 1 for
descriptive statistics): number of full-time employees; number of
full-time reporters and editors; number of beats or departments;63 and
number of promotions needed for reporter to become top editor.64 The
second dimension is ownership structure: whether the newspaper was owned
by chain or group (no=0; yes=1); whether public ownership was possible
(no=0; yes=1); whether the newspaper was a legally incorporated business
(no=0; yes=1); and whether the newspaper was controlled by one family
or
individual (no=1; yes=0). The third dimension is the presence of
rules and
procedures, composed of two measures: whether the newspaper has "its own
formal, written code of ethics" and whether the newspaper has "its own
employee handbook of rules and procedures." The fourth dimension is
rationality, operationalized as the amount of perceived importance top
management places on "finding the most efficient way to solve
problems."65
And the fifth dimension is staff expertise, measured by a question which
asked whether "reporters normally need a bachelor's degree to be
considered
for employment at your newspaper (no=0; yes=1)"
-------------------------------
Insert Table 2 about here
--------------------------------
A research assistant coded two major items from the editorials and the
letters to the editor in each issue. The first was focus:
An editorial or letter to the editor or commentary is defined as local if
the main topic of the article focuses primarily on issues,
activities or
matters that affect or concern the local community, which, for
purposes
here, is defined as the newspaper's primary market area. This
includes the
city and county in which the newspaper is located and, in the case of
metropolitan dailies, the surrounding suburbs and urbanized areas.
City
hall, county government, and regional planning groups, for example,
are
local groups or organizations. However, state government, because
its a
ctivities focus on the entire state and not just a local area, is
defined
as a nonlocal group, even if the statehouse is located in the
community the
newspaper serves. Any issue or matter associated with the federal
government is nonlocal. If it is not clear whether the focus is local or
nonlocal, choose local.
To assess the reliability of this measure, this researcher coded the
content of 64 editorials and subjected the findings to a reliability
statistical test. Intercoder reliability on this measure was 96%. The
coefficient of reliability used was the ratio of coding agreements to
the
total number of coding decisions (CR=[Number of Agreements x 2] / [N1
+N2]).[14]
The second content item coded was critical evaluation:
Is the main emphasis of the editorial or letter to the editor negative,
positive or neutral in terms of the way it evaluates an action, rule,
law,
decision, position, value, idea, ideology, custom or practice
associated
with a mainstream individual or group?
Positive is defined as content that is commending, applauding, approving or
admiring. Negative is defined as content that is faulting, blaming,
censuring, or disapproving. If a story contains content that is
positive
and negative toward the mainstream entity, the evaluation should be
based
on an overall assessment of the article (i.e., was it more positive
than
negative?). Stories, editorials, etc. that contain neither content
that is
positive or negative or content that appears to be equally balanced should
be coded neutral. If it does not deal with content that is critical or
praiseworthy, then code it neutral. When a story contains no
reference to
a mainstream group, value, etc., it should be coded not applicable.
Mainstream is defined as an individual or group that is associated with
local, state or federal government (e.g., city hall, police, schools,
Congress, the President, courts, colleges, mayor, governor, state
agencies,
city council members, etc.), the two main political parties (Democrat and
Republican), private business and corporations, or mainstream
churches
(Catholic, mainstream protestant or Jewish).
The vast majority of editorials dealt with issues surrounding the actions
or activities of government. Intercoder reliability for the
four-category
measure of critical evaluation (positive, neutral, negative, not
applicable) was not extremely high (64% agreement). To rectify this
problem, the critical evaluation measure was dichotomized into critical
vs.
noncritical. This produced an intercoder reliability coefficient of 86%.
The source of the editorial was also coded:
Who wrote the editorial D someone from the local newspaper staff (usually
identified as staff writer or in the case of editorials, no byline)
or
nonstaff, which includes content generated by wire services,
syndicates,
other newspapers, or special to the paper?
The reliability coefficient here was 100 percent.
The mean number of editorials produced over the two-day period was 2.64.
Most of those are staff-generated (66%) but only minority are about
local
issues (24%). About 4 of 10 (40%) contain content that is critical of
mainstream groups or authorities. The typical newspaper also published
about 5 editorials over the two-day period, most of which were about
local
issues (70%). About 4 of 10 (37%) also contain content that is
critical of
mainstream groups or authorities. Additional descriptive statistics for
the content measures are presented in Table 2.
Findings
H1: The data support the first hypothesis, which expected that the more a
newspaper exhibits the characteristics of the corporate form of
organization, the greater the number of editorials and letters to the
editor. Table 3 shows that the zero-order correlation between the
corporate structure index and number of editorials is .42 (p<.01). The
corresponding correlation for letters to the editor is .49 (p<.01).
Three
of the five corporate dimensions D structural complexity, rules and
procedures, and hire college graduates D are significantly related to the
corporate structure index. The correlation for the ownership
structure
dimension is positive but not statistically significant. All five
dimensions are significantly related to the number of letters to the
editor. For illustrative purposes, the last column in Table 3 also shows
the correlation between chain ownership (0=independent; 1=chain) and
the
editorial-page measures. In neither case are the relationships
significant.
-------------------------------
Insert Table 3 about here
--------------------------------
H2: The data also support the second hypothesis: The more a newspaper
exhibits the characteristics of the corporate form of organization,
the
greater the number and proportion of staff-generated editorials. Table
3
shows that the zero-order correlation between the corporate newspaper
index
and number of staff-generated editorials is .43 (p<.01); the correlation
for proportion of staff-generated editorials is .33 (p<.01). Three of
the
five corporate dimensions D structural complexity, rules and
procedures,
and staff expertise (hire college graduates for reporting positions) D
are
significantly related to both of the staff-generated measures. The
correlations for chain ownership show that it is not significantly related
to either of the staff-generated measures.
H3: The data provide partial support for the third hypothesis, which
expected that corporate organizations would publish a greater number and
proportion of editorials and letters to the editor about local issues.
Table 2 shows that the correlation between the corporate newspaper
index
and number of local editorials is .22 (p<.05). Two of the five
dimensions
D structural complexity and hire college graduates D are significantly
related to the number of local editorials measure. The corporate index
is
not significantly related to the proportion of editorials (r=.09,
p>.05),
but the structural complexity dimension is (r=.19, p<.01; i.e.,
larger,
more complex organizations write more local editorials). The
correlations
between the corporate newspaper index and letters to the editor are
statistically significant and consistent with expectations (r=.45 for
number and r=.19 for proportion). The chain ownership measure is not
significantly related to either of the local editorial measures.
H4: The data support the fourth hypothesis D that the more a newspaper
exhibits the characteristics of the corporate form of organization,
the
greater the number and proportion of editorials critical of mainstream
groups or sources. The correlation between the corporate index and
number
of critical editorials is moderately strong (r=.33, p<.01; see Table
3).
Four of the five dimensions D structural complexity, ownership
structure,
rules, and staff expertise D are significantly related to this
measure.
The correlation between the corporate index and proportion of critical
editorials also is statistically significant (r=.20, p<.01), but only
structural complexity and staff expertise are related to this measure.
The
relationships between the corporate index and number of critical letters
to the editor is quite strong (r=.52, p<.01). All five of the
dimensions
are significantly related to corporate structure. The correlation
between
the corporate index and proportion of critical editorials is
moderately
strong (r=.32, p<.01). Four of the five corporate dimensions are
related
(the exception is rational decision-making). Again, the chain
ownership
measure by itself is not related to either of the critical
editorial-page
measures.
Summary and Discussion
Contrary to the critical model, this study hypothesized that corporate
newspapers would publish more staff-generated editorials, more
editorials
about local issues, and more editorials that are critical of
mainstream
groups than entrepreneurial newspapers. Corporate newspapers publish
more
staff-generated editorials and more editorials about local issues
because,
it was argued, they are larger, more structurally complex, and have
more
staff and resources. Corporate newspapers write more editorials that
are
critical of mainstream groups because they are more likely to be
located in
pluralistic communities that contain more criticism of dominant groups and
value systems and because they are more insulated from local political
pressures. On the latter point, the owners and managers of corporate
newspapers are less likely to grow up in the community their newspaper
serves, are more likely to work at the newspaper for a shorter period of
time, are oriented to the larger corporation, not the local community,
and
are more heavily influenced by professional norms and values, which
place a
higher premium on truth and criticism than on local parochial interests.
Findings support nearly all of the hypotheses. Corporate newspapers
publish a larger number and proportion of editorials and letters to the
editor, a larger number and proportion of staff-generated editorials, a
larger number of editorials that deal with local issues, a larger
number
and proportion of letters to the editor that deal with local issues,
and D
most importantly D a larger number and proportion of editorials and
letters to the editor that are critical of mainstream groups and
institutions. Consistent with much of the literature, this study found no
relationship between editorial-page content and chain ownership.
Chain
ownership, it was argued, is too crude of a measure of organizational
structure to be useful by itself.
The strength of the structural model presented here is that it helps to
account for social change and the role that mass media often play in
promoting such change. As social systems become more pluralistic, news
media within those systems become more critical of traditional ways and
established institutions. Media reflect to some degree the diversity
of
the communities they serve, and increasing role specialization and
professionalization, by-products of community and organizational growth,
insulate journalists from political pressures. The increased level of
criticism that emerges from these structural forces contributes to
discourse that places increased pressure on existing institutions to
change.
Despite the strengths of the structural model tested here, one shortcoming
is that it cannot determine whether social change is occurring quickly
enough to be of optimal benefit to society and/or challenging or
alternative groups. Many critics would counter that even if corporate
newspapers are more critical of mainstream groups than entrepreneurial
newspapers, the editorial criticism they offer produces little in the
way
of meaningful social change. Such critics may be right. But in light
of
the findings of this study, the validity of such arguments will need
to
depend more on value judgments about meaningful social change than on
the
notion that corporate newspaper structure leads to less diversity on
the
editorial pages. ENDNOTES
Table 1. Summary of Literature on Editorial Page Vigor*
Studies which find
corporate newspapers
are less vigorous
Studies which find
few differences or
have mixed findings
Studies which find
corporate newspapers
are more vigorous
Wackman, Gillmor,
Gaziano, Dennis 1975
Thrift 1977
St. Dizier 1986
Borstel 1956
Grotta 1971
Wagenberg & Soderlund 1975
Ghiglione 1984
Gaziano 1984
Busterna & Hanson 1990
Hale 1988
ASNE 1979
Goodman 1982
Daugherty 1983
Akhavan-Majid, Rife, &
Gopinath 1991
Wilhoit & Drew 1991
Akhavan-Majid &
Boudreau 1994
*Vigor is defined here as content that is more critical of established
authorities or more diverse, or structural conditions that give journalists
more
autonomy or more ability to criticize. Table 2. Descriptive
Statistics for
Measures
Mean Std Dev
Minimum Maximum
N
Independent Variables
1. Overall corporate index (2+3+4+5+6)_ .00 1.00
-2.65 3.18
172
2. Structural complexity index_ .00 1.00
-1.22 5.08
188
y Number of promotions reporter needs
to become top editor 3.14 1.22
1 8
190
y Number of beats employing full-time reporters 5.30 4.25
0 18
192
y Number of full-time reporters/editors 38.25 63.38
2 500
190
y Number of full-time employees 184.62 472.67
2 7,000
189
3. Ownership structure index_ .00 1.00
-1.58 1.54
180
y Proportion that are incorporated businesses .81 .39
0 1
190
y Proportion publicly owned .29 .46
0 1
190
y Proportion in which one family/individual does not
own 50 percent interest in newspaper .63 .45
0 1
184
y Proportion owned by chain/group .67 .47
0 1
192
4. Rules and procedures index_ .00 1.00
-1.27 1.27
190
y Proportion that have their own formal, written code
of ethics .33 .47
0 1
190
y Proportion that have their own employee handbook of
rules and procedures .68 .47
0 1
191
5. Rational decision making (mean on 7-point scale) 4.77 1.72
1 7
191
(Standardized)_ .00 1.00
-2.19 1.29
191
6. Proportion requiring reporters to have a college degree .73 .44
.00 1.00
185
(Standardized)_ .00 1.00
-1.64 .61
185
Editorial Page Measures (for 2 issues)
1. Total number of editorials 2.64 1.84
0 8
198
2. Number of staff-generated editorials 2.16 1.92
0 8
198
3. Proportion of editorials that are staff-generated .66 .43
0 1.00
198
4. Number of editorials about local issues .72 .92
0 4
198
5. Proportion of editorials about local issues .24 .31
0 1.00
198
6. Number of editorials that are critical of
mainstream groups/institutions/sources 1.31 1.34
0 6
198
7. Proportion of editorials that are critical of
mainstream groups/institutions/sources .40 .36
0 1.00
198
8. Total number of letters to the editor 5.09 4.62
0 18
198
9. Total number of letters about local issues 4.22 3.84
0 15
198
10. Proportion of editorials about local issues .70 .39
0 1.00
198
11. Number of editorials that are critical of
mainstream groups/institutions/sources 2.34 2.53
0 12
198
12. Proportion of editorials that are critical of
mainstream groups/institutions/sources .37 .31
0 1.00
198
_Measures are standardized with mean=0 and SD=1. Table 3. Zero-Order
Correlations Between
Corporate Structure and Editorial Page Measures
Rational Hire
Corporate Structural Ownership Rules &
Decision- College
Chain
Index_ Complexity Structure Procedures
Making Grads
Ownership__
(172) (188) (180) (190)
(191) (185)
(192)
Number of Editorials .42** .59** .11 .24** -.03
.27** -.01
Number of Letters
to the Editor .49** .61** .18* .29**
.13* .23** .01
Staff-Generated
Editorials
Number .43** .68** .09 .30** -.06
.24** -.03
Proportion .33** .45** .07 .23* -.01
.23** .05
Editorials About
Local Issues
Number .22* .41** .04 .10 -.09
.15* .06
Proportion .09 .19** .01 -.01 -.04
.11 .08
Letters About
Local Issues
Number .45** .50** .17** .26**
-.17** .22** .01
Proportion .19** .09 .05 .05
.13* .30** -.05
Editorials that
are critical of
mainstream groups
Number .33** .46** .16* .17** -.08
.16* -.02
Proportion .20** .24** .09 .10 -.02
.13* .05
Letters that
are critical of
mainstream groups
Number .52** .61** .20* .29**
.18** .21** .01
Proportion .32** .26** .12* .15* .09
.27** .10
*p<.05; **.01
_Corporate Index = structural complexity + ownership structure + rules &
procedures +
rational decision making + hire college graduates
__Chain ownership (chain=1; independent=0) is one component of the ownership
structure dimension but is
shown separately here for illustrative purposes.
December 23, 1994
Dr. John Soloski, editor
Journalism Monographs
School of Journalism and Mass Communication
University of Iowa
Iowa City, Iowa 52242
Dear Professor Soloski:
Enclosed are three copies of a manuscript, titled "Corporate Newspaper
Structure and Editorial Page Vigor," that I would like to submit for
possible publication in Journalism Monographs.
Thank you for the feedback on my previous submission ("Corporate Newspaper
Structure and Organizational Goals"). The reviewers offered some good
ideas and advice for improving the manuscript. I intend to shorten it
and
submit to another journal.
Thank you for considering this request.
Sincerely,
David Pearce Demers
Assistant Professor
[1] Some studies use more than two newspapers as a cu
t-off criteria, while
others used size of chain.
[2] Demers, "Structura
l Pluralism, Intermedia Competition, and the Growth
of the Co
rporate Newspaper in the United States," argues that chain
ow
nership is no longer a good measure of corporate complexity in
cross-sectional studies at the organizational level because it is so wide
ly
diffused in the industry. However, chain ownership appears to be a go
od
measure in longitudinal studies at the system level (i.e.,
where chain
ownership is measured as percent of total newspa
pers owned by groups).
[3] Demers, "Corporate Newspapers Structure and Or
ganizational Goals."
[4] For reviews of the literature, see David Pearce
Demers and Daniel B.
Wackman, "Effect of Chain Ownership on N
ewspaper Management Goals,"
Newspaper Research Journal 9 (Win
ter 1988): 59-68, and David Pearce Demers,
"Corporate Structure and Empha
sis on Profits and Product Quality at U.S.
Daily Newspapers,"
Journalism Quarterly 68 (Spring/Summer 1991): 15-26.
[5] Structural plur
alism may be defined as the number and variety of
groups and
organizations in a social system. See Phillip J. Tichenor,
G
eorge A. Donohue and Clarice N. Olien, Community Conflict and the Press
(Beverly Hills, CA: Sage, 1980), 16.
[6] Wilson has shown that
as population increases, heterogeneity
increases, partly bec
ause the probability that there will be enough people
(i.e.,
critical mass) to form a group that exhibits alternative views
increases. Thomas C. Wilson, "Community Population Size and Social
Heterogeneity: An Empirical Test," American Journal of Sociology 9
1 (March
1986): 1154-1169.
[7] The American Society of News
paper Editors' code of ethics contains the
following dictum: "The Americ
an press was made free not just to inform or
just to serve as a forum for
debate but also to bring an independent
scrutiny to bear on
the forces of power in the society, including the
conduct of
official power at all levels of government." Quoted in Philip
Meyer, Ethical Journalism (New York: Longman, 1987), 21.
[8] Empirical
support for this conclusion can be found in George Albert
Gla
dney, "How Editors and Readers Rank and Rate the Importance of 18
Traditional Standards of Newspaper Excellence" (Paper delivered at AEJ
MC
annual meeting, Atlanta, 1994). Gladney's study shows tha
t editors at
larger newspapers (size being one measure of com
plexity) are much more
likely to say newspapers should be agg
ressive in their reporting and
willing to hire top, professio
nal staffers.
[9] Douglas Anderson, "How Managing Editors View and Deal W
ith Ethical
Issues," Journalism Quarterly 64 (Summer/Autumn 1
987): 341-345.
[10] Meyer, Ethical Journalism, 28-29, provides empirical
support for
this. He reports the results of a national ASNE
survey which shows that
journalists at large newspapers are m
uch less likely than those at smaller
newspapers to be restra
ined when reporting on controversial topics.
[11] Theodore L. Glasser, "P
rofessionalism and the Derision of Diversity:
The Case of the
Education of Journalists," Journal of Communication 42
(Spri
ng 1992): 134. Glasser's main argument is that higher education
journalism programs reify professional standards and fail to teach thei
r
students how professionalism promotes social order, a point
with which I
wholeheartedly agree.
[12] See, e.g., J. Herb
ert Altschull, Agents of Power: The Role of the
News Media in
Human Affairs (New York: Longman, 1984); W. Lance Bennett,
N
ews: The Politics of Illusion, 2nd ed. (New York: Longman, 1988); Stuart
Ewen, Captains of Consciousness: Advertising and the Social Roo
ts of the
Consumer Culture (New York: McGraw-Hill, 1976); Mar
k Fishman, Manufacturing
the News (Austin: University of Texas Press, 198
0); Herbert J. Gans,
Deciding What's News (New York: Vintage,
1979); Todd Gitlin, The Whole
World Is Watching: Mass Media
in the Making and Unmaking of the New Left
(Berkeley: Univers
ity of California Press, 1980); Morris Janowitz, The
Communi
ty Press in an Urban Setting (Glencoe, IL: Free Press, 1952); David
L. Paletz and Robert M. Entman, Media Power Politics (New York: The
Free
Press, 1981); Tichenor, Donohue and Olien, Community Co
nflict and the Press
; Gaye Tuchman, Making News (New York: Fr
ee Press, 1978); Gaye Tuchman,
"Mass Media Institutions," in
Neil Smelser, ed. Handbook of Sociology
(Newbury Park, CA: Sa
ge, 1988).
[13] In some cases, the newspapers would send tear sheets only
after
payment of a small charge.
[14] O. R. Holsti, Conten
t Analysis for the Social Sciences and Humanities
(Reading, Mass.: Addiso
n-Wesley), 140.
|
