   

3% “HHIWHHIWWWIIHIHIWHHIWIINIIIH!NW

 

LIBRARY
Michigan. State

llhi! 'AVH:“ :
UIIIVCIDIly

 

 

 

This is to certify that the
dissertation entitled

PERCEPTIONS OF DEVIANCE AND DEVIANT GROUPS IN THE
NEWS: NEWSPAPER COVERAGE OF ARAB AMERICANS PRE-
AND POST-9/ 11

presented by

MIRON VAROUHAKIS

has been accepted towards fulﬁllment
of the requirements for the

Ph.D. degree in Communication Arts &
Sciences - Media & Information
Studies

 

 

fatal/W A [3101.”th

\ Major Professors Signature
(Deal? 1‘" ﬂ 07

Date

MSU is an Afﬁnnative Action/Equal Opportunity Employer

----v-»----—.---o-v--.-o-v—.—.-.—--—---I----.-.--—o-o-—-c-o--u-a-o-u-.-o----u-o-c----u-u-u----—.-u-.-n—

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due. '
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 K:/Proilec&Pros/CIRCIDateDue.indd

 

PERCEPTIONS OF DEVIANCE AND DEVIANT GROUPS IN THE NEWS:
NEWSPAPER COVERAGE OF ARAB AMERICANS PRE- AND POST-9/ll
By

Miron Varouhakis

A DISSERTATION

Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Communication Arts & Sciences -
Media & Information Studies

2009

ABSTRACT

PERCEPTIONS OF DEVIANCE AND DEVIANT GROUPS IN THE NEWS:
NEWSPAPER COVERAGE OF ARAB AMERICANS PRE- AND POST-9/ll

By

Miron Varouhakis

The ﬁndings Show that deviance does play some role in the news coverage of
Arab-Americans, but the role is not as strong as theory would predict. Contrary to the
prediction by the theories of inﬂuence on mass media content by Shoemaker and Reese,
the study found that an increase in the US. public’s perceived deviance of Arab-
Americans was not correlated with increased stereotyping of them by eight major metro
newspapers in the ﬁve years that followed the Sept. 11, 2001 terrorist attacks. The
ﬁndings also provided only moderate support to the hypothesis that the more deviant the
US. public perceives Arab-Americans the more coverage they will receive by the news
media — also an argument of the Shoemaker and Reese theory. The study also found that
although there was an increase in coverage of Arab-Americans after the terrorist attacks
of 9/11 — as the Shoemaker and Reese theory predicted — that increase declined in the
following years. The ﬁndings also show that the majority of the coverage of Arab-
Americans remained balanced in tone in the ﬁve years after 9/11, while the use of local

Arab community sources more than doubled in the years aﬁer 9/11.

AKNOWLEDGEMENTS
In publishing this dissertation I would like to take note of all those who helped me
in making it a reality. I would like to specially thank Dr. Steve Lacy, my dissertation
committee chair and adviser, for all his guidance and unparallel support throughout the
process. Special thanks also go to the rest of the members of my dissertation committee:
Dr. Keith Adler, Dr. Lucinda Davenport, and Dr. Fred Fico. Their guidance and steadfast

support were instrumental to the completion of the project.

iii

TABLE OF CONTENTS

LIST OF TABLES .................................................................................. V
INTRODUCTION .................................................................................... 1
THEORETICAL FRAMEWORK ................................................................. 6
LITERATURE REVIEW ........................................................................... 15
HYPOTHESES & RESEARCH QUESTIONS ................................................ 24
RESULTS ............................................................................................ 46
CONCLUSIONS ................................................................................... 85

APPENDICES ....................................................................................... 90
Content Analysis Protocol ............................................................... 90

ENDNOTES ...................................................................................... 101

iv

LIST OF TABLES

TABLE 1 Public Opinion about Deviance of Arab-Americans
and News Coverage by Location of Newspapers ............................................... 48

TABLE 2 Public Opinion about the Deviance of Arab-Americans
and Presence of Stereotypes in News Coverage by Location of Newspapers ............... 52

TABLE 3 Percentage of Stories with Stereotypes in Coverage of
Arab-Americans by Region from 1996 to 2006 ................................................. 54

TABLE 4 Percentage of Stories with Stereotypes in Coverage of
Arab-Americans by Bach Newspaper from 1996 to 2006 ..................................... 55

TABLE 5 Total Number of Stories about Arab-Americans
Published by Bach Newspaper from 1996 to 2006 ............................................. 58

TABLE 6 Type of Stories about Arab-Americans Published
by Region Five Years Prior and Five Years after 9/11 ......................................... 60

TABLE 7 Type of Stories about Arab—Americans
Published by Bach Newspaper Five Years Prior to 9/11 ....................................... 61

TABLE 8 Type of Stories about Arab-Americans
Published by Bach Newspaper Five Years after 9/1 1 .......................................... 62

TABLE 9 Geographical Focus of Stories about Arab—Americans
Published by Region Five Years Prior and Five Years after 9/11 ............................ 64

TABLE 10 Topical Focus of Stories about Arab-Americans
Published by Region Five Years Prior and Five Years aﬁer 9/11 ............................ 66

TABLE 11 Type of Sources in Stories about Arab-Americans
Published by Region Five Years Prior and Five Years after 9/11 ............................ 69

TABLE 12 Tone toward Focus of Stories about Arab-Americans
Published by Region Five Years Prior and Five Years after 9/11 ............................ 73

INTRODUCTION

The perceived deviance of minority groups by the American society and the news
coverage of those minorities by US news media is an issue that is ftmdamental to both
sociology and communication scholarship. Sociologists have investigated the concept of
deviance since 1895 when Durkheim published The Rules of Sociological Method.1 In
communication studies deviance is investigated both in terms of its effects on content as
an external ideological inﬂuence2 and within the context of media effects.3

The concept of deviance, although deﬁned since the times of Durkheim, has
evolved. Today, sociology scholars talk about negative deviance, rate busting, deviance
admiration, and positive deviance.4 In communication, Shoemaker and Reese deﬁne
deviance not as an individual belief system but a societal-level phenomenon. The media,
as agents of social control, do not screen out deviant ideas but rather portray them in a
way calculated to underscore their deviance.5 Thus, deviance in communication is
deﬁned negatively and describes underconformity, or nonconformity, by a minority
group — or an individual — within society. Scholars have also argued that minority groups
— political or otherwise — are by deﬁnition inherently deviant.6 This of course doesn’t
mean that scholars deﬁne minorities negatively, but rather underline that by nature those
in minorities have views that are not shared by the majority.

Minorities are an integral part of the American society, often called a ‘melting
pot’ or ‘salad bowl’ as references to the coexistence of its many minorities. These
minorities play an integral role in the social and cultural fabric of the United States and
merit the attention of communication scholars. Keever et al. in 1997 examined the news

coverage of Native, African, Hispanic and Asian Americans, as well as Paciﬁc Islanders

from 1934 to 1996 and found that the minorities in the study had been subjected to biased
and distorted representations in the news media.7 The authors note that there were
signiﬁcant gaps in the coverage, which was deﬁned as degrading and having reinforced
racial attitudes and practices that existed in the US. society for decades.8 In this case,
according to the ﬁndings of the study, the public perceptions about speciﬁc minority
groups were adopted by the news media, which in turn echoed those perceptions in its
content — one of the core arguments in the Shoemaker and Reese Theories of Inﬂuence on
Mass Media Content.9

This study comes as a continuation of that scholarship by investigating the news
treatment of another important minority — that of Arab-Americans — through the deviance
lens. Over 3 million Americans trace their ancestry to an Arab country,10 a minority
which after the terrorist attacks of Sept. 11, 2001 has been the target of discrimination
and hate crimes. The FBI recorded a seventeen-fold increase in anti-Muslim crimes
nationwide during 2001, while Human Rights Watch in its 2002 annual report cited more
than 2,000 September ll-related backlash incidents.ll

In relation to deviance, the Arab-American community would be seen as deviant
by the mainstream for three reasons: a) they are non-traditional immigrants because they
are from the Middle East rather than Europe, Africa or Asia, which are the homelands for
most US. immigrants in the earlier years, b) The group is perceived, whether rightly or
wrongly, as being Muslim, which makes them “deviant,” and c) the 9/11 attacks led some
people to assume all Arabs are enemies of the US. In other words, the Arab-American

community is likely to be seen as deviant for multiple reasons.

This study investigates the association between general perception of deviance
and news media content by examining the news coverage of the two largest Arab-
American communities — those in California and Michigan — by state and regional
newspapers before and after the terrorist attacks of Sept. 11, 2001. The investigation is
ﬂamed around the Theories of Inﬂuence on Mass Media Content by Shoemaker and
Reese, and speciﬁcally the Inﬂuence of Ideology which includes deviance as a factor.‘2
JUST IF ICA TION

Studying societal-level inﬂuences on news coverage about perceptions of
deviance is important because it can advance our understanding of the inﬂuence society
has on news media content. It will also expand our understanding about the role of the
news media in the coverage of minorities and how they contribute to existing ideological
perceptions with society about speciﬁc minorities that are deemed as deviant.

Shoemaker and Reese deﬁne ideology as a symbolic mechanism that serves as a
cohesive and integrated force in society, and they argue that media play a role in
reinforcing this ideology.l3 Indeed the signiﬁcance of the news media’s role in society
was highlighted in a strongly worded Hutchins Commission Report in 1947, in which a
group of mostly academics called on American journalists to take on more responsibility
to society. The Commission criticized the news media for coverage that was
characterized by “meaninglessness, ﬂatness, distortion, and the perpetuation of
misunderstanding.”l4 The Commission made ﬁve recommendations — branded as a code
of social responsibility for the press in a free society — where it also calls for “the

projection of a representative picture of the constituent groups in the society.”15

Keever showed that indeed American news coverage between 1934 and 1996 had
subjected minorities to biased and distorted representations.‘6 The literature by
communication scholars on deviance and news has found that news coverage does play a
role in the way audiences perceive deviant groups — political or otherwise — and that it
serves as a predictor of newsworthiness. '7 Shoemaker and Reese note that
communication is an essential part of deﬁning deviance and argue that the media do not
just convey the labels created by others. They also make their own decisions about tone,
emphasis, placement, and portrayal. ‘8

Thus, studying societal-level inﬂuences on news coverage about perceptions of
deviance is important because it opens a path for communication scholars in explaining,
controlling, and predicting ideological inﬂuences about deviance on news content at
societal level, as well as the role of the press in perpetuating those ideological views
about deviance or in creating new ones.

Investigating the correlation between the general perception of deviance between
people in the American society and news content through the news coverage of the Arab-
American community offers several advantages.

First, the Arab-American community has the unique position as having multiple
deviance elements: Arab-Americans are non-traditional immigrants; they are all seen as
Muslims in the eyes of the American public; and after 9/11 some Americans see them as
a threat to the security of the US. Although Muslim Arab-Americans makeup only a
small fraction of the entire Arab-American population19 this study examines Arab-
Americans as one group. This decision is backed by the fact that for the longest time the

American public has been assuming that most, if not all, Arab-Americans are Muslims.20

Such misleading assumptions existed even among American journalists, who in their
reporting have been using interchangeable the terms “Muslim,” “Arab,” and “Islam,”
perpetuating that way the misconceptions that have been harbored by the American
public.21 So widespread were these practices that the Detroit Free Press was prompted to
publish a joumalist’s guide about Arab-Americans in which it explained among others
that most Arab-Americans am not Muslim, but Christian.22

In terms of the views of the American public about Islam, several polls conducted
after the Sept. 11 terror attacks showed that religion was a serious concern among the
American people when it came to terrorism and conﬂict. A March 2002 poll conducted
by the Pew Forum on Religion & Public Life found that 65% of those polled said that
religion plays “a great deal” or “a fair amount” in causing war and conﬂict.23 A second
poll by the Pew Forum on Religion & Public Life the following year found that 44% of
Americans who were polled believe that Islam encourages violence, a sharp increase
from the 25% a year earlier.24 These views of the American public about religion, and
particularly Islam, formed a “religious basis” of American responses to removing
Saddam Hussein, invading Iraq, and perceiving Islam as a violent religion.25 These views
continued to grow well into 2006, when a poll by Washington Post and ABC News found
that 46% of Americans view Islam negatively.26 These views are also very important
because religion remains as one of the core institutions in human culture. According to
2004 estimates, more than 2.1 billion people in the world identify themselves as
Christians and more than 1.3 billion identify themselves as Muslims.27

Second, using the Arab-American community to investigate the relationship

between the general perception of deviance among the American society and news

content is also advantageous because Arab-Americans are one of the largest ethnic
communities in the United States — based on the 2000 US. Census ﬁgures they are
ranked seventh among all ethnic minorities.28 California and Michigan are home to the
two largest Arab American communities in the country.29 As such, this study also
contributes to a rather slim body of communication literature that focuses on a speciﬁc
ethnic group and its portrayal in newspapers operating in their communities. As national
polls have shown, almost half of the American public has a negative view of Islam and
this study wants to investigate the role of the media in the overall portrayal of Arab-
Americans, Muslims, and Islam.

Finally, the decision to examine local newspapers instead of other news medium
was based on existing scholarship that has demonstrated that local newspapers remain

important to people as a source of local news.30

THEORETICAL FRAMEWORK

According to the Theories of Inﬂuence on Mass Media Content by Shoemaker and
Reese,31 there are ﬁve levels of inﬂuences on media content: a) individual media
workers, b) media routines, c) organizational, d) actors and institutions outside of media
organizations, and e) societal ideology. They present these inﬂuences in the form of a
hierarchical model, with the “individual media workers” being that central layer — with
the least inﬂuence on content — and societal “ideology” at the outer layer. The layer of
ideological inﬂuences includes also inﬂuences of public perceptions of deviance on news
media content. This study focuses mainly on the inﬂuences of the public’s perceptions of

deviance about individuals, groups and events on news media content. In doing so, it will

also explore the role of extramedia inﬂuences — included in the “outside of media
organizations” inﬂuences — and particularly the targeted audience of a news media outlet,
has on news content in the context of deviance. Shoemaker and Reese note that the layer
of extramedia inﬂuences includes economically and politically powerful sources;
government sources, interest groups and public relations ﬁrms, advertisers, and targeted
audiences.
Escher

The term ideology is said to have been coined by French enlightenment
philosopher Destutt de Tracy in 179632 as a reference to his “science of ideas.” Since then
the concept has been evolving through the works of Marx, Freud, Nietzsche, Lukas,
Gramsci, Adorno, Bourdieu, among others.33 In 1977 Britain’s leading post-war cultural
historian and theorist, Raymond Williams deﬁned ideology as a “relatively formal and
articulated system of meanings, values and beliefs, of a kind that can be abstracted as a
‘worldview’ or a ‘class outlook.”34

Other deﬁnitions of ideology include those by Becker and Williams. Becker notes
that ideology governs our perception of the world and ourselves, acting as an integrated
set of references or frames.3S According to Williams, ideology is somewhat formal and
articulated system of values, meanings, and beliefs that can serve as a ‘class outlook.’36
In 1976 Gouldner noted that “ideology assumes special importance as a symbolic
mechanism through which interests of these diverse social strata may be integrated;
through the sharing of it the several dominant strata are enabled to make compatible

responses to changing social conditions.”37

In their theory, Shoemaker and Reese note that over the years the concept of
ideology has been broadened to include feminist theory, deconstructionism, semiotics,
and other post-modem approaches.38 Instead Shoemaker and Reese try to provide some
key concepts and reﬂect the kinds of questions that concern communication scholars that
study media at the ideological level. Thus, Shoemaker and Reese deﬁne ideology as a
symbolic mechanism that serves as a cohesive and integrated force in society, and they
argue that media play a role in propagating this ideology.39

In terms of the process, they say that ideology works through existing values and
thus should not be viewed as an alien belief system that is simply being implanted on a
culture.40 Using the case of the United States as an example, Shoemaker and Reese note
that there is a fundamental belief in the values of the capitalist economic system (e. g.
private ownership, proﬁt-driven entrepreneurship, free market, etc.). That system, they
say, is intertwined with the Protestant ethic and the value of individual achievement and a
set of political values that are centered around a liberal democracy.4|

They also argue that ideology is not an individual belief system, but rather a
societal-level phenomenon — keeping in line with the European tradition of media studies
in which ideology is considered a total structure, as opposed to a system of individual
attitudes and values.42 As such, they say that process of inﬂuencing news media content
is instigated not by an individual but as a class, transcending any one organization,
industry, or place.43 An example of such process could be a citizen-based boycott of a
Speciﬁc newspaper over its coverage of a speciﬁc issue or affair that they disapproved,

where in turn the newspaper decides to apologize in order not to lose its readership.

Hall in 1989 argued that the media have the ability to ‘deﬂne situations’ and label
groups and individuals as deviant by maintaining the ideological boundaries of society.44
Shoemaker and Reese adopt this argument, noting that one of the key functions
performed by media is to maintain boundaries in a culture, and that in order to integrate
societal interests, some views and values must be deﬁned as within the bounds of
acceptability, whereas others area read out of legitimacy.45 Deviance is one of those
ideological constructs that society negotiates and communicates through the news media.

This approach to ideology by Shoemaker and Reese, though, is more of a critical
studies one and its application to a social sciences model comes with some limitations —
including an immeasurable level of vagueness. Social sciences traditionally deal with
variance, but Shoemaker and Reese do not discuss ideology in terms of variance. They
rather present it as a total structure, arguing that ideology does not vary much within a
society. They argue that journalists will not use objective routines, such as balance, when
subjects are outside the area of legitimate controversy and in the areas of consensus or
deviance.46 In other words they argue that ideology has a blanket effect on news media
coverage and being more powerful of a factor than organization inﬂuences, such as the
owners of media outlets, the editors, individual media workers '(such as the reporters),
etc. This proposed lack of variance in the Shoemaker and Reese model makes it difficult
to test in a social sciences context But societal-level perceptions of deviance — which
according to Shoemaker and Reese exist within the outer circle of societal ideology — can

be measured because it varies, and thus can be studied within a social sciences context.

Deviance

Like ideology, the concept of deviance has a long history and has been a long-
standing interest to social scientists. The sociology of deviance grew to include different
perspectives about deviance, including the structural-functionalist perspective and
symbolic interactionism, as well as a constellation of theories, including the
neautralization theory, anomie, labeling theory, control theory, and conﬂict theory.
Structural functionalism, also known as “social systems theory,” grew out of a notion
introduced by Comte and Spencer, under which a social entity, such as an organization or
a whole society, can be viewed as an organism.47 Under this perspective, Durkheim’s
“anemic” theory suggests that crimes and their punishment provide an opportunity for
institutions to reafﬁrm society’s values, including those about deviance.48 On the other
hand, the symbolic interactionism perspective focuses on the micro level — whereas the
structural functionalism perspective focuses on the macro level such as the institutions.”
Under symbolic interactionism, most interactions revolve around the process of
individuals reaching common understanding through the use of language and other such
systems, including the media.” Thus, under this school of thought deviance is both
learned and sidelined through communication.

In their theory Shoemaker and Reese adopt the symbolic interactionist
perspective, and they view deviance not as an unchanging condition but as continually
being deﬁned and renegotiated as the participants interact with each other symbolically.5 '
They note that the media are continually coping with new ideas, reafﬁrming social
norms, and redrawing or deﬁning boundaries, thus making communication an essential

part of deﬁning deviance.52

10

Communication scholars who have studied deviance have supported the position that
news media content does inﬂuence audiences’ perceptions about deviance, while several
dimensions of deviance have been found to play a key role in the newsworthiness criteria
of news media, including news values of controversy, sensationalism, prominence, and
unusualness.53 Shoemaker also found that those political groups perceived as deviant by
newspaper editors were typically given less favorable treatment, and although they were
not given less prominent attention, their legitimacy was more likely to be questioned. 5"
Moreover, Shoemaker et al. in 1984 found that world events covered by the US. media
were more deviant than those not covered, in the sense that those events (such as
terrorism) threatened the status quo in the country in which they occurred.55 Similarly,
covered events often conveyed normative deviance — that is, they would have broken
American norms had they occurred in the United States.56

As suggested by the Shoemaker research, the news media do not screen out
deviant ideas but rather portray them in a way calculated to underscore their deviance.57
In short, deviance is related to ideology as it involves people’s perceptions of how
individuals, groups and events vary ﬂom the accepted ideology. Thus, because deviance
is based on perceptions, it can vary within a society and can be measured as a social ~
sciences variable.

Wells argues that deviance has been deﬁned in three ways: a) under the normative
way, behavior is deviant when it breaks social rules or norms and it is the society that
determines who is deviant, b) under the labeling perspective, a group is deviant when and
because someone calls it deviant, and c) behavior is consciously deviant to the extent that

a person or group is aware that his or her actions are in some sense wrong or socially

11

disapproved.58 Shoemaker and Reese argue that in the context of deviance and social
change, the media act as a key control mechanism in society, and that the normal is
reafﬁrmed by being presented routinely and in juxtaposition to the deviant, which
competes at the boundaries for attention. 5 9 They say that as agents of social control, the
media reafﬁrm the ideological status quo by ridiculing deviant ideas, adding that political
groups that have been shut out of the media spotlight may become even more shrill and
radical and thus continuing the original deviant label.60

The idea of boundaries is based on a model developed by Hallin in 1986 by which
he explained the ways in which news media reinforce ideological boundaries by dividing
the journalistic world into three spheres: a) sphere of consensus, b) sphere of legitimate
controversy, and c) sphere of deviance.61 In the model, the three spheres are represented
into three layers, with the sphere of consensus at the center and the sphere of deviance at
the outer layer.62 Hallin says that at the sphere of deviance the reporter’s role is to expose,
condemn, or exclude from the public agenda those who violate or challenge the political

COI'ISCI‘ISIIS.63

Minority Groups

Just like ideology and deviance, the term minority has also long historical roots.
The term minority was originally emerged in the context of the rise of nationalism and
the nation-state in the late-eighteenth—century and early-nineteenth-century Europe.64 It
was used to characterize national or ethnic groups that had become subordinate to the
peoples of another national or ethnic group through the shifts in political boundaries.65
The term was adopted in the United States after World War I, primarily in response to the

publicity given to issues involving European minorities that were addressed in the

12

negotiations that ended the war."6 However, in the United States the term developed a
different meaning than in Europe, as the minority peoples of the United States were
ﬁghting for status within the larger society when the minorities of Europe were mainly
ﬁghting for independence ﬂom it.67 Gleason contented that the way in which the social
scientiﬁc concept of minority evolved in the United States reﬂected implicit American
cultural and political assumptions about the nature of minorities — i.e., that minorities are
the consequence of exclusion by majorities from the mainstream of American society and
that, once tolerated and accepted as equals, minorities will ultimately disappear.68

American social scientists have generally used the terms majority and minority to
refer to systems of structured social inequality in which racial and ethnic criteria play a
critical role in the system’s ranking system.69 The term minority has been applied with
greatest frequency to subordinate groups characterized by hereditary membership and
endogamy - racial, caste, and ethnic groupings.70 Thus a minority group can do deﬁned
as any group that is assigned an inferior status in society, that is, any group that has less
than its proportionate share of wealth, power, or social status, and which are often
discriminated against by those in the majority.71 As Keever et al. found in their 1997
examination of the news coverage of Native, Aﬂican, Hispanic, Asian Americans, and
Paciﬁc Islanders, ﬂom 1934 to 1996, minorities had been subjected to biased and
distorted representations in the news media.72

Minorities in society are not only assigned an inferior status in society, but
according to Moscovici a minority group is by deﬁnition deviant.73

This study connects the concepts of ideology, deviance, and minorities through

the Shoemaker and Reese Theories of Inﬂuence on Mass Media Content and speciﬁcally

13

the Inﬂuence of Ideology. One of the main hypotheses in Inﬂuence of Ideology is that the
more deviant people or events are, the more likely they are to be included in media
content and the more likely they are to be stereotyped.74 The hypotheses bring together
the concepts of ideology (at the societal level), deviance (ideologically deﬁned by
society), and minority (by deﬁnition any minority group is deviant ﬂom the majority and
it is the majority that ideologically deﬁnes what and who is deviant).

In this study, the Arab-American community represents the minority group, while
a series of national surveys by the Pew Forum on Religion & Public Life that were
conducted between 2001 and 2005 provide the perspective of the American public about
the deviance of Arab-Americans and Islam. The term stereotype used in the hypothesis
by Shoemaker and Reese will be explored in this study by using stereotypes relevant to
Arab—Americans that have been identiﬁed in existing scholarship. For example, after the
terrorist attacks of 9/11 the US. pubic may perceive that Islam is violent and that
Muslims were anti-American. These stereotypes of deviant nature can be identiﬁed in the
manifest content of newspaper coverage of Muslim-Americans, and they are also
measurable because they vary.

In summary, Shoemaker and Reese contend that in their ﬁve layers of inﬂuence of
content — where individual media workers are at the center and societal-level ideology is
at the outer layer - the individual media worker has the least inﬂuence on media content.
Speciﬁcally, they note that the ideological layer of inﬂuence is the most powerful one as
all the processes taking place at the lower layers are thought to be working toward an
ideologically related pattern of messages and on behalf of the higher power centers of

society.75 This may not always be the case, though. A 1955 study by Breed on social

14

control in newsrooms in the United States found ﬁve cases where staffers (or individual
media workers) manage to bypass content policy that was instituted by the media
organization."5 The ﬁve cases highlighted by Breed were: a) Vagueness in policies gives
ﬂexibility and autonomy to staffers; b) On some issues staffers can have superior
knowledge than their executives; c) Staffers can bypass executive orders of dropping a
story by “planting” the story in another news outlet via a ﬂiendly colleague in that outlet;
(1) Staffer-initiated stories offer greater autonomy to the reporter; and e) Staffers with
“star” status can transgress policy more easily.77 Breed’s study show that individual
media workers may have a lot more inﬂuence on content than Shoemaker and Reese

suggest.

LITERATURE REVIEW

Deviance and News

The body of communication literature that has examined deviance and news is
lengthy78 but the majority of the studies focus on the impact of news coverage on
audience perceptions of deviance. Studies that investigate news media content effects on
audience perceptions of deviance span a period of several decades, and they have shown
that news coverage does play a role in the way audiences perceive deviant minority
groups - political or otherwise. In 1982 Shoemaker conducted two experiments in which
subjects were shown mock newspapers stories about non-mainstream political parties.
The ﬁndings Showed that subjects' evaluations of a party's legitimacy increased in
favorability when they read a positive, serious story, while it declined when they read a

negative ridiculing story.79

15

These earlier ﬁndings were reinforced later when McLeod in 1995 conducted an
experiment that examined television news coverage’s impact on audience’s perceptions
of social protests. Subjects were shown a television news story about a 1987 protest by an
anarchist group in Minneapolis.80 The ﬁndings showed that subjects who saw the one-
Sided news story — in which the opinions of the protesters were excluded - were
signiﬁcantly less critical of the police officers and more critical of the protesters, while
they also tended to identify less with the protesters.81 The ﬁndings ﬂom the early 19808
and mid-19903 were further buttressed in 2008 when Lee conducted an online experiment
using different levels of deviance in news stories. During the experiment, some 111
participants browsed the Web with their behaviors recorded by a Web-tracking program.
The ﬁndings showed that deviance was a strong predictor of story selection while
personal involvement interfered with the inﬂuence of story deviance.82

Communication scholars have also investigated the newsworthiness factor of
deviance and studies have shown that deviance serves as a predictor of newsworthiness.
In 1984 a study by Shoemaker found that the news media report less favorably on deviant
political groups but not less prominently.83 This prominence, or newsworthiness, of
deviance was also investigated by Weirnann et al. in 1991, who examined the
newsworthiness of international terrorism. The ﬁndings suggest that in the case of
terrorism, deviance as newsworthiness can be regarded as the guiding principle of
coverage.84

Six years later Pritchard et al. sought to develop a more sturdy theoretical

explanation for journalists’ decisions about which crimes (or deviant act) to highlight and

which to ignore. The authors proposed that four forms of deviance (normative deviance,

16

statistical deviance, status deviance, and cultural deviance) account for much of the
variation in decisions about crime news.85 The deviance-based explanations for crime
news were tested through a comprehensive content analysis of two Milwaukee
newspapers and in-depth interviews with journalists at the two media outlets.86 The
ﬁndings showed that the newsworthiness of a homicide is enhanced when Whites are
suspects or victims, males are suspects, and victims are females, children, or senior
citizens.87 They concluded that status deviance and cultural deviance are important
components of newsworthiness and that statistical deviance (unusualness) may be much
less important than commonly assumed.88 That same year, Breen examined deviance as a
trigger for intermedia agenda setting by investigating changes in trends of news media
reports of deviant acts by members of the clergy after a negative triggering event of child
sexual abuse between 1991 and 1995.89 The ﬁndings of a content analysis of some 235
stories suggested strong media agenda-setting effects of the negative triggering events or
the subsequent coverage of the clergy in general.90

One area where communication scholars have not focused their investigation
enough is the ideological inﬂuence of society’s perception of deviance about minorities
on news content.

Murray et al. in 2008 investigated the extent to which protesters opposing the
2003 Iraq War were successﬁll at securing positive coverage in the British press
immediately before and during the invasion of Iraq.91 Britain was selected because the
war was highly controversial in the United Kingdom, generating domestic opposition and
a widely supported anti-war movement.92 The ﬁndings of a content analysis showed that,

although anti-war protesters received more favourable than unfavourable coverage prior

17

to the war, once the war got under way, a ‘support our boys’ consensus led to the
narrowing of the ‘sphere of legitimate controversy’ with the anti-war movement relegated
to a 'Sphere of deviance.’93 The ﬁndings also showed that elite-led protest was more
successful at inﬂuencing newspaper debate than grassroots protest.94 Overall, the authors
note that the ﬁndings highlight the problems protest movements have in securing positive
media representation during war.95 Shoemaker and Reese said that the normal is
reafﬁrmed by being presented routinely and in juxtaposition, which competes at the

boundaries for attention.96

Newsﬁ and Arab-Americﬂ

The body of communication literature that investigates the news coverage of
Arabs and Arab-Americans by US. news media is rather thin but grong in recent
years. The earliest studies found that Arabs and Arab-Americans have been cast as
deviant. One of the earliest and most notable studies on Arab images in the West was by
Said in 1979.97 In his book Orientalism, Said concluded that the Western world had
created Arabs as a racial other, portrayed as dangerous, emotionally volatile, and
backwards.98 Studies that followed in the 19808 found that the prevalent stereotypical
images of Arab Americans were those of terrorists, hijackers, and religious
fundamentalists.”

More than a decade later, though, Gavrilos found that news features published in
The New York Times, USA TODA Y, and the Washington Post during the period of the
1991 Gulf War about Arab-American reactions to the war had sympathetic portrayals of
Arab-Americans who feared the loss of Arab lives and the threat of racism while

highlighting their steadfast loyalty to America.100 But the study used one story per

18

newspaper, which placed serious limitations to the conclusions and to any inferences to
the national medialm

The consequences of 9/11 on Arab-Americans have also become the focus of
scholarly research. Domke et al. in 2003 examined 345 news stories ﬂom 26 national and
statewide newspapers that were published ﬁve months prior and after the 9/11.102
Focusing on the manner in which “racial proﬁling” was talked about by government and
societal leaders, nongovernmental groups, and average citizens, the ﬁndings showed that
Arab Americans spoke ﬂom a more favorable position of status than Aﬂican Americans,
and that racial minorities may face a situation in which they de facto “compete” with
other minorities for Space in news coverage. '03

Another study by Weston that same year examined qualitatively 195 news stories
published in the Chicago Tribune and the Detroit Free Press between June 1, 2001 and
Oct. 11, 2001. He found that the two newspapers resisted stereotypes and discrimination,
describing Arab-Americans as doubly victimized and loyal patriotic members of the
community. '04 The study, though, was limited by the timeﬂarne used in the investigation
and the number and types of newspapers used in the sample.

Two years later, a study by Salaita contradicted those ﬁndings by examining the
effects of 9/11 on the Arab-American community, with an emphasis on how notions of
patriotism have altered both American and Arab-American life.105 Offering analyses of
pedagogy, ethnic studies, xenophobia, racism, and stereotype in order to highlight the
complexities of the interaction between Americans of Arab origin and other domestic

ethnic groups, Salaita concluded that 9/11 did not actually alter American attitudes

toward Arab-Americans, but rather it reinforced attitudes both positive and negative that

19

had existed for decades.“ Salaita noted that 9/11 simply offered racists and xenophobes
a rhetorical trope that could legitimize their views, while it oﬁ‘ered multiculturalists more
reasons to promote inclusionary ideasm
This study continues that scholarship but moves away ﬂom the approach that
wants researchers examining content for presence or absence of speciﬁc characteristics

and seeks to explore if characteristics that relate to deviance correlate with the prevalent

perceptions in the American public.

Arab-Americans

Americans of Arab descent have been part of the country’s unique multicultural
tapestry for over a century.108 The ﬁrst signiﬁcant wave of Arab immigrants began slowly
in the 18703, reached a zenith between 1900 and 1913, and continued at a reduced level
until 1924.")9 Some historian estimate that more than 100,000 Arabs left their homelands
for American by 1914, with 90% of them being Christians ﬂ'om what it was then the
Ottoman Province of Greater Syria — a region that included present-day Jordan, Lebanon,
Syria, and Israel and the occupied Palestinian territories.1 ‘0 Others place the number
Arabs who came to American even higher at 250,000.1 '1 The inﬂux of Arabs into the
United States halted in the 19208 after the passing of the National Origins Act in 1924,
which restricted immigration to 2% or about 150,000 per year.”2 The period between
1920 and 19503 Arab immigrants started to gradually become assimilated to the
American culture, but that also meant a decline in Spoken Arabic and the Arabic press.1 '3

The second wave of Arab immigrants to America came in the 19503 and 19603, triggered

by political and economic upheaval in many Arab countries, such as Egypt, Palestine,

20

Syria, Iraq, and Yemen.1 ‘4 This second wave of Arab immigrants was very different from
the ﬁrst because 60% of them were Muslim.115 Between 1965 and 1992 it is estimated
that more than 400,000 Arab immigrants came to America.116 This second wave of Arab
immigrants also became the catalyst in the reawakening of an ethnic consciousness
among Arab-Americans, which also came to be known in the community as “Arab
Renaissance.”l ‘7

Today, at least 3.5 million Americans are of Arab descent, while more than 80%
of them are US citizensm Lebanese-Americans constitute the greater part (39%) of the
total Arab-Americans, while others include Egyptians (12%), Syrians (12%), and
Palestinian (6%).119 In terms of employment trends among Arab-Americans are very
similar to the national average, with 65% of Arab-American adults in the labor force and
a 5% unemployment rate. 120 Most Arab-Americans (88%) work in the private sector,
while 73% of them are employed in managerial, professional, technical, sales, or
administrative ﬁelds.'2' At the same time, the median income for Arab-American
households in 1999 was $47,000, compared with $42,000 for all households in the US. at
the time.122 In terms of religion, 35% are Roman/Eastern Catholic, 24% Muslim, 18%
Eastern Orthodox, 10% Protestant, and 13% Other Religion/No Afﬁliation. '23 A study by
Kulczycki and Lobo in 2001 that examined the changing demographical characteristics
and socioeconomic proﬁle of Arab-Americans concluded that although still small in size
relative to many other US. ethnic groups, the Arab-American population may gain more
public recognition if current demographic trends hold and a positive assimilation path is
maintained.124

Arab-Americans in California

21

The Arab population of California is the largest in the United States with more
than 700,000 Arab-Americans living there.125 The Arab community of California is one
of the most vibrant in the country, having grown by more than 34% between 1990 and
2000.126 At the same time the Arab/Chaldean ancestry population of California grew by
an average of more than 57,000 in each decade.127 The majority of California’s Arab-
American community is of Lebanese origin (24%).'28 Other large groups included those
ofArab origin (17%), Egyptian origin (14%), Syrian (9%), and Palestinian (7%).129 The
Arabs of California have settled in 56 out of all 58 counties in the state, but some 66% of
them live in the greater Los Angeles area, San Diego, and the seven northern counties
around the San Francisco Bay.130 Speciﬁcally, 32% of California’s Arabs reside in the
Los Angeles County, 12% in the San Diego County, another 12% in Orange County, 6%
in Santa Clara County, and 4% in the San Bemardino County.131 The remaining 44% is
spread in smaller communities throughout Califomia132

In spite the fact that this is the largest Arab-American presence in a Single state it
is still statistically less signiﬁcant when compared to other ethnic groups in the state.‘33
For example, in the Los Angeles Metropolitan Area where there is the largest presence of
Arab-Americans in the state, the largest ethnic groups are Whites (55%), Asian (10%),
Aﬂican-American (8%), Native-American (1%), and Paciﬁc Islanders (0.5%).I34 Thus,
although some 34% of Arab-Americans of California reside in the Los Angles

Metropolitan Area, they only account for 0.01% of the population.135

22

Arab-Americans in Michigan

Arab immigrants have had a presence in Michigan for more than a century, and
the community’s economic and cultural prominence - measured in thousands of
businesses and hundreds of mosques, churches and social clubs — make it uniquely
vibrant.136 Michigan is home to the highest concentration of Arab-Americans in any US.
state, but the size of the Arab-American community varies depending on the source, and
can be anything ﬂom 1 15,000137 to 200,000138 and even 490,000.”9 The Arab-American
community is represented in 82 out of 83 counties in Michigan, with more than 80% of
the state’s population residing in the three Detroit metro counties of Macomb, Oakland,
and Wayne.140 Roughly one third of the city of Dearborn claim some Arab heritage.141 In
terms of the countries of origin, 36% are Lebanese, 23% Assyrian/Chaldean/Syriac, 16%
Arab/Arabic, 7% Iraqi, 6% Syrian, 3% Palestinian, 2% Egyptian, 2% Jordanian, and 6%
Other.142 In terms of religion, the Arab-American community of Detroit is divided into
two religious groups, Muslim and Christian.143 Similarly with the case of California,
although this is the second largest Arab-American presence in a state, it is still
statistically less signiﬁcant when compared to other ethnic groups in the state.144 For
example, in the Detroit Metropolitan Areawhere the entire Arab-American community of
the state is residing, the largest ethnic groups are Whites (73%), Blacks (21%), Hispanic
(3%), and Asian (2.5%).145 Thus, Arab-Americans, although 100% of them live in the
area, they account for approximately 0.01% of the population — approximately an equal
size with that of the Los Angeles Metropolitan Aream’

But in Spite the vagueness of its size, the Arab-American community of Michigan,

and particularly that in the metro Detroit area, have demonstrated a high degree of

23

success in business and have proven to be inﬂuential in municipal and state politics.”7

Arabs have been coming to Detroit since the late nineteenth century, and they continue to
arrive in thousands each year.148 In the book “Arab Detroit: From Margin to
Mainstream,” Abraham and Shryock note that the ﬁrst Arab immigrants arrived in
Detroit even before the auto industry begun to boom:

“In 1900 the Detroit Free Press, February 6, ran a story on “Detroit’s Colony of
Syrians.” The reporter describes a small, exotic community of 75-100 people, most of
them Maronite Christians ﬂom Lebanon. These early immigrants were peddlers and
shopkeepers, not industrial workers. By 1916, however, there were already 555 Syrian
men employed in Henry Ford’s factories alone, many of them newly-arrived Muslims. By
1930, there were nine thousand Arabic-speaking Detroiters: almost six thousand were
Syrian, and the rest were a mix of Palestinians, Yemenis, and Iraqi Chaldeans.”

Today’s AraboAmerican community of Detroit evolved out of that existing and
growing population, and gradually started to integrate into its gulfs that newly arriving
immigrants who had come to America as refugees of war — Shia ﬂ'om Lebanon and Iraq,
and victims of political oppression, economic hardship, and population displacement —
Iraqi Catholics, Palestinians and Yemenis.149 According to scholars, Arab-Americans

tOday are ($116 Of Michigan’s largeSt, faStCSt growing minority poplﬂations.150

HYPOTHESES & RESEARCH QUESTIONS

The study addresses a total of four hypotheses and four research questions, with

the majority of them aimed at testing the Shoemaker and Reese Theories of Inﬂuence on

24

Mass Media Content and speciﬁcally the societal-level inﬂuences on news content about
deviance. 151

One of the central arguments in the Shoemaker and Reese theory has been that the
more deviant a group or events involving the group are the more coverage the group will
receive from the news media. '52 The hypothesis contends that public views about
perceived deviant events and groups- expressed in the public domain through collective
ways such as national polls — inﬂuence the perceptions of journalists and editors about
those events and groups and thus the news content that they produce about them. For
example, the US. public may perceive that Mexicans who are trying to cross the border
into the United States are deviant (seen as a threat to national security) and express their
perception in national surveys. In turn, journalists and editors after reviewing those polls
decide to increase coverage of Mexicans.

This argument is explored through the case of Arab—Americans. Speciﬁcally, with
the use of data ﬂom a series of national surveys by the Pew F orurn on Religion & Public
Life that were conducted between 2001 and 2005, and three measures of deviance were
used about Arab-Americans. After establishing the general perceptions of the American
public about the deviance of Arab-Americans, the study carried out a quantitative content
analysis of eight large US. daily newspapers in order to test the Shoemaker and Reese
argument by exploring the ﬁrst hypothesis:

H1: The more deviant Arab-Americans or events involving Arab-Americans are

perceived to be by the American public, the more likely they are to be covered in the

news by the Los Angeles Times, the San Diego Union-Tribune, the Detroit Free Press,

25

the Detroit News, the Oregonian, the Milwaukee Journal Sentinel, the Denver Post, and
the Minneapolis Star Tribune.

Another central argument in the Shoemaker and Reese theory has been that the
more deviant a group or an event involving that group is the more likely it is to be
stereotyped by the press.153 This argument is at the core of their theoretical argument that
has news media content being inﬂuenced ideologically by society, particularly when it
comes to deviance and deviant groups. Similarly with their previous hypothesis, this
contends that public views about perceived deviance of a group - expressed in the public
domain through collective ways such as national polls — will lead journalists and editors
at news media outlets to deem that Arab-Americans are worthy of additional coverage. In
this case, the journalists and editors fall back on stereotypes about the speciﬁc group that
they have acquired though mediated and interpersonal communications. Again, the
argument was explored through the case of Arab-Americans. The deviance of Arab-
Americans was established with the Pew surveys, while a quantitative content analysis of
eight large US. daily newspapers allows a test of the Shoemaker and Reese argument
through the second hypothesis:

H2: The more deviant Arab-Americans or events involving Arab-Americans are
perceived to be by the American public the more likely they are to be stereotyped in the
news coverage of the Los Angeles Times, the San Diego Union-Tribune, the Detroit Free
Press, the Detroit News, the Oregonian, the Milwaukee Journal Sentinel, the Denver
Post, and the Minneapolis Star Tribune.

The Shoemaker and Reese proposition on “extramedia inﬂuences” argues that the

more a media organization promotes itself within a target audience, the more its content

26

will reﬂect the interests of that audience.154 This hypothesis contends that the
marketplace (environment or audience) within which a news media outlet operates
inﬂuences the news content that it produces. For example, if a newspaper operates within
a community where the majority of the people are Jewish, the journalists and editors will
be inclined to be more sensitive to issues related to Judaism so that they will not alienate
their readers and even lose them. In this case, the marketplace’s inﬂuence on news media
content is stronger than the societal-level ideological inﬂuences. The study explores the
assumption that the extramedia inﬂuences are stronger than ideological inﬂuences
through the case of Arab-Americans, whose deviance was established with the Pew
surveys. The assumption is tested by a content analysis of eight large US. daily
newspapers — four of which are based in areas with large Arab-American communities
and four in areas with insigniﬁcant or nonexistent Arab-American presence — and through
the third hypothesis:

H3: The Los Angeles Times, the San Diego Union-Tribune, the Detroit Free
Press, and the Detroit News will be less likely to include stereotypes about Arab—
Americans in their coverage with the Oregonian, the Milwaukee Journal Sentinel, the
Denver Post, and the Minneapolis Star Tribune.

The Shoemaker and Reese theory also argues that the more deviant an event is the
more coverage it will generate.155 Again, the study explores this argument through the
case of Arab-Americans, whose deviance will be established with the Pew surveys, and
by using the terrorist attacks of 9/11 as the case of the deviant event. The argument is

tested by a content analysis of eight large US. daily newspapers — four of which are

27

based in areas with large Arab-American communities and four in areas with
insigniﬁcant or nonexistent Arab-American presence — and through the fourth hypothesis:

H4: News coverage of Arab-Americans by the Los Angeles Times, the San Diego
Union-Tribune, the Detroit Free Press, the Detroit News, the Oregonian, the Milwaukee
Journal Sentinel, the Denver Post, and the Minneapolis Star Tribune will increase after
the terrorist attacks of 9/11, an event that was deviant by nature.

The study also explores four research questions that sought to address issues
related to the nature of coverage of a deviant group by offering detailed insights about the
structure of the coverage. The study measures nature of coverage based on four indicators
(sources, focus, type of story, and tone). Speciﬁcally, the research questions seek to
address issues of depth of coverage, whether the focus of the stories was local, national,
or international, the type of sources that appeared more often in those stories, and ﬁnally
the tone of the stories. Data generated by these questions allows the study to offer the
kind of detailed analysis that is not limited only to the volume of the coverage or
differences between certain newspapers, but also explores different aspects that are
related to the nature of the news coverage.

The questions that aimed at addressing those issues are the following:

RQl: How many feature and in-depth stories did the Los Angeles Times, the San
Diego Union-Tribune, the Detroit Free Press, the Detroit News, the Oregonian, the
Milwaukee Journal Sentinel, the Denver Post, and the Minneapolis Star Tribune carry
about Arab-Americans pre- and post-9/11?

“Feature stories” and “in-depth stories” are measured by a single variable Type of

Story that has three categories: hard news story; feature story; and in-depth story. It is

28

argued that feature stories are much lengthier and detailed in their analysis in comparison
to hard news — published a newspaper. Features also require a lot more time and
resources than hard news, an indication that the news medium is showing a stronger
interest in an event or issue. In-depth stories differ ﬂom hard news and feature stories.
For the purposes of the study they include news analysis and investigative stories. In-
depth stories differ ﬂom hard news stories in that they go beyond pure facts about a
timely event and offer expert analysis of what the event means for present time and the
future. They also differ from features because they are not human interest stories. “In-
dep ” stories offer analysis and projection about current events through analysis ﬂ'om
experts on the topic

RQ2: What was the focus of the Arab-related stories of the Los Angeles Times,
the San Diego Union-Tribune, the Detroit Free Press, the Detroit News, the Oregonian,
the Milwaukee Journal Sentinel, the Denver Post, and the Minneapolis Star Tribune pre-
and post-9/ 1 1?

“Focus” is measured by two variables, Geographical Focus and Topical Focus.
Geographical Focus has three dimensions: a) local community story, b) localization of
national affairs, and c) localization of international affairs. The speciﬁc three categories
of the “geographical focus” variable were selected because they allowed the study to
identify if the stories about Arab-Americans were dealing with issues deriving ﬂom the
local Arab-American community or at the national and international levels. The second
variable, Topical Focus, has also three dimensions: a) local Arab community affairs, b)

discrimination/attacks against Arabs, and c) religion. The speciﬁc three categories of the

29

“topical focus” were selected because they allowed the study to identify if some key
topics were reﬂected in the coverage and how much.

RQ3: Which types of sources appeared most often in the stories about Arab-
Americans or Arab affairs published in the Los Angeles Times, the San Diego Union-
Tribune, the Detroit Free Press, the Detroit News, the Oregonian, the Milwaukee Journal
Sentinel, the Denver Post, and the Minneapolis Star Tribune pre- and post-9/11?

“Sources” is measured by a total of six variables: local Arab community
members; national Arab community members; local government ofﬁcials; federal
government ofﬁcials; law enforcement ofﬁcials; and foreign government officials. The
six “sources” variables were selected because they allowed the study to identify who
were the most ﬂequently used sources in the stories about Arab-Americans, as well as to
see how much voice local Arab community members and national Arab community
members had in the stories dealing with their communities compared to ofﬁcial
government entities at the local, federal, and international level. A higher presence of
local Arab community members in the stories would be an indicator that the newspapers
made an effort to include local Arab views in the stories that were related about them or
issues about them, while a higher use of government sources would indicate that the
newspapers relied more heavily on official sources and not paying enough attention to the
views of the local Arab community. Each variable was compared with the other ﬁve
based on the frequency it appears in the stories.

RQ4: What was the tone toward the focus of the Arab—related stories published by

the Los Angeles Times, the San Diego Union-Tribune, the Detroit Free Press, the Detroit

30

News, the Oregonian, the Milwaukee Journal Sentinel, the Denver Post, and the
Minneapolis Star Tribune pre- and post-9/11?

“Tone” will be measured by one variable: Tone toward Focus. The variable has
four categories: positive toward focus; balanced toward focus; negative toward focus; and
story is entirely neutral. This last variable — used in the “2006 News Availability Project”
as part of a national study for the Project for Excellence in Journalism — was selected
because it will allow the study to identify if the newspaper coverage of Arab-Americans
was positive, balanced, negative, or neutral, and if there was a change in the coverage

tone after 9/11.

METHODS

Newspapm

This study uses a quantitative content analysis of eight dailies - two from
Califomia, two ﬂ‘om Michigan, one ﬂom Oregon, one from Wisconsin, one ﬂ'om
Minnesota, and one from Colorado. Speciﬁcally, the newspapers selected for the content
analysis are the Los Angeles Times, the San Diego Union-Tribune, the Detroit Free
Press, the Detroit News, the Oregonian, the Milwaukee Journal Sentinel, the Denver
Post, and the Minneapolis Star Tribune. The eight newspapers were selected for two
primary reasons: a) their location in relation to the concentration, or lack off, of Arab-
Americans in the six states, and b) their circulation ﬁgures.

A) Location. Speciﬁcally, in California the two dailies represent an area where
more than 42% of the state’s Arab-American population resides!“5 In Michigan the two

dailies are published in Detroit Metropolitan Area that covers an area where more than

31

82% of the state’s Arab population resides.15 7 In both states, although the Arab-American
presence is the highest in the country, statistically the Arab—Americans are far less vibrant
in their presence than other ethnic groups.158 Also, in the areas where the four dailies are
located in the two states Arab-Americans are of similar size, about 0.01% of the
metropolitan population.159 The other four newspapers ﬂom Colorado, Minnesota,
Oregon, and Wisconsin were selected because they don’t have any signiﬁcant presence of
Arab-American communities,‘60 thus allowing to test for differences in coverage with
those dailies that serve areas with considerably highly concentrated Arab-American
communities. Moreover, the four newspapers were selected for the geographic location,
with the Oregonian and the Denver Post representing states in the greater West region
with no signiﬁcant Arab-American presence, and the Milwaukee Journal Sentinel and
Minneapolis Star Tribune representing states in the greater Midwest region with no
signiﬁcant Arab-American community. The underlying mechanism is that newspapers
that do not cater to large Arab-American communities will be more likely to treat them as
deviant than newspapers that cater to large Arab-American communities. This relates to
the Shoemaker and Reese hypothesis of “extramedia inﬂuences,” which argues that the
more a media organization promotes itself within a target audience, the more its content
will reﬂect the interests of that audience.161
B. Circulation. Finally, the speciﬁc newspapers were selected because of their
large audience. Speciﬁcally, their high circulation numbers are: the Los Angeles Times
(775,766), the Minneapolis Star Tribune (358,887), the Detroit Free Press (345,861), the
Oregonian (310,803), the San Diego Union-Tribune (304,334), the Denver Post

(258,696), the Milwaukee Journal Sentinel (230,781), and the Detroit News (201,432)."52

32

These circulation ﬁgures place these eight dailies in the top 50 within the Top 100 US.
dailies.“S3 These newspapers not only have some of the highest circulation numbers in the
country, but also in their state — thus reaching larger audiences than any other newspaper
in their state.

Sampling

The archives of the eight dailies were accessed through Lexis/Nexis, NewsBank,
Factiva, and online archives of newspapers.

The sample for the content analysis was drawn by using an electronic keyword
search to ﬁnd all news stories, about “Arab-Americans” that were published by the eight
dailies between Jan. 1, 1996 and Dec. 31, 2006. For the purposes of the study “news” is
deﬁned as information about recent events as they are reported and published by the eight
daily newspapers. The “news” can be in three different formats: a) Hard news, b)
Features, and c) In-depth Stories.

“Hard news” involve stories that report on a speciﬁc timely event or series of
events, with a focus on what, when, where, and why, and which commonly appear on the
front page or any other pages in the news section. “Features” these are also known as
human-interest stories because they usually focus on prominent or unusual events and
individuals. They differ ﬂom hard news stories in that the topic is not necessarily timely
and that they are much lengthier. In newspapers this type of stories usually appears in a
“Features” section, but in some instances they don’t have their own section and thus
appear under the “Local” or “State” sections. “In-depth Stories,” include news analysis
and investigative stories. They differ from hard news stories in that they go beyond pure

facts about a timely event and offer expert analysis of what the event means for present

33

time and the future. They also differ ﬂorn features because they are not human interest
stories. News analysis stories offer analysis and projection about current events through
analysis ﬂom experts on the topic. They are commonly identiﬁed by a “News Analysis”
logo either next to the headline or inside the story. Any news story, feature, and in-depth
that mentions “Arab-American” were selected for the ﬁnal sample. Of those, only stories
with two or more mentions of “Arab-American” was included for the analysis in an effort
to eliminate stories that just have a passing mention about Arab-Americans. Investigative
stories also require a lot of resources and time ﬂom the part of the newspaper, and they
can be either in the form of a single story or a series of stories. They can usually be
identiﬁed by an acknowledgement within the story, such as “an investigation by [title of
newspaper],” or even on the headline.

The years of the sampling starts on Jan. 1, 1996 and ends on Dec. 31, 2006. The
time ﬂame was selected based on the rationale that the ﬁve year period prior to 9/11
allows to record characteristics of the news coverage prior to the terrorist attacks, while
the ﬁve year period aﬂer 9/11 allows to record any changes to the news coverage. In this
case, the terrorist attacks of Sept. 11, 2001 act as the treatment in what is a quasi-
experimental study.

The study used four different samples in order to best address the hypotheses and
research questions.

“Full Census Sample”: This ﬁrst sample was a census and covered all of stories
about Arab-Americans that were published by eight dailies between 1996 and 2006. The
sample had an N=2,010 (Los Angeles Times 322; San Diego Union Tribune 186;

Oregonian 85; Detroit Free Press 539; Detroit News 484; Milwaukee Journal Sentinel

34

212; Denver Post 81; and Minneapolis Star Tribune 10]). This sample was used to
answer H4, which seeks to test if the amount of coverage of Arab-Americans by the eight
dailies increased after 9/11, covering a period of over 10 years (Jan. 1, 1996 to Sept. 10,
2001 and Sept. 12, 2001 and Dec. 31, 2006). The day of September 11, 2001was
intentionally excluded ﬂom the sample as any coverage of Arab-Americans that day
would have been exclusively about that event.

“Partial Census Sample”: This second sample was also a census and covered the
total number of stories about Arab-Americans published by the eight newspapers during
ﬁve three-month periods that followed each national Pew survey. The sample had an
N=241 (Los Angeles Times 44; San Diego Union Tribune 43; Oregonian 15; Detroit Free
Press 58; Detroit News 39; Milwaukee Journal Sentinel 33; Denver Post 5; and
Minneapolis Star Tribune 14). This sample was used to answer H1 and H2. H1 seeks to
test if there is a correlation between the amount of coverage of Arab-Americans and their
perceived deviance by the US. public. H2 seeks to test if there is a correlation between
the presence of stereotypes in the coverage of Arab-Americans and their perceived
deviance by the US. public.

“Main Rapdom Sample”: This third sample was random and was drawn ﬂom the
census sample ofN=2,010 stories that were published by the eight dailies during the
1996-2006 period. The sample had an N=500 (Los Angeles Times 85; San Diego Union
Tribune 45; Oregonian 25; Detroit Free Press 130; Detroit News 114; Milwaukee
Journal Sentinel 50; Denver Post 26; and Minneapolis Star Tribune 25). This sample was
used to answer H3, which seeks to test if all eight newspapers used stereotypes against

Arab-Americans equally in their coverage from over a decade (Jan. 1, 1996 to Sept. 10,

35

2001 and Sept. 12, 2001 to Dec. 31, 2006). The sample was proportionate to the
percentage of total stories about Arab-Americans published by the eight newspapers
being studied between 1996 and 2006. The use of a random sample in this case was
deemed appropriate as the sheer volume of stories in the census exceeded the 2,000 thus
making a random sample more efﬁcient.

“Large Random Sample”: This fourth sample is also random and was drawn ﬂom
the census sample of N=2,010 stories that were published by the eight newspapers during
the 1996-2006 period. The sample had an N=554. The pre-9/11 sample had an N=l6l
and the post-9/11 sample had an N=393. The distribution of stories per newspaper in the
pre—9/ 11 sample was: the Los Angeles Times (33), the San Diego Union-Tribune (10), the
Detroit Free Press (54), the Detroit News (30), the Oregonian (10), the Milwaukee
Journal Sentinel (10), the Denver Post (9), and the Minneapolis Star Tribune (5). The
distribution of stories per newspaper in the post-9/11 sample was: the Los Angeles Times
(53), the San Diego Union-Tribune (45), the Detroit Free Press (84), the Detroit News
(85), the Oregonian (25), the Milwaukee Journal Sentinel (50), the Denver Post (26), and
the Minneapolis Star Tribune (25).

The sample was used to answer all four research questions. All four research
questions seek to investigate different variables during the entire 1996-2006 period. The
use of a random sample in this case was deemed appropriate as the sheer volume of
stories in the census exceeded the 2,000 thus making a random sample more efﬁcient.
Due to a very low volume of pre-9/11 stories in the census sample of N=2,010 by four
dailies — the Oregonian (85), the Milwaukee Journal Sentinel (212), the Denver Post

(81), and the Minneapolis Star Tribune (101) — oversampling of those dailies was deemed

36

necessary. Speciﬁcally, the Oregonian had only 22 of its 85 stories published prior to
2001. Random sampling based on its proportion in the census sample (85 stories out of
2,010) could not have produced a large enough random sample that would be needed for
a valid analysis. Similarly, the Denver Post and the Minneapolis Star Tribune had each
published only 15 stories about Arab-Americans between 1996 and 2001, while the
leaukee Journal Sentinel had published only 40 stories. The larger random samples
from the four dailies (the Oregonian, 35 stories, or 41%; the Milwaukee Journal Sentinel,
60 stories, 28%; the Denver Post, 35 stories, or 43%, and the Minneapolis Star Tribune,
30 stories, or 30%) were deemed more appropriate as they allowed for greater validity in
the analysis by raising the number of pre-9/ 11 stories by the four dailies in the random
sample. For the two dailies (the Denver Post and the Oregonian) that had less than 100
stories in the census sample, a random sample of about 40% was deemed efficient, while
for the other two dailies (the Milwaukee Journal Sentinel and the Minneapolis Star
Tribune) that had over 100 stories in the census sample, a random sample of about 30%
was deemed adequate.

Two coders were trained for the content analysis and a two-person coder
reliability assessment were conducted using a random sample of 10% (50 stories of the
total N=500 “Main Random Sample”) and was tested based on a Scott’s Pi statistical
measurement. It was decided to use the Scott’s Pi test because it corrects the coefficient
of reliability for agreement due to chance, and achieves by measuring the percentages of
both observed and expected agreements. Also, the decision to use a 10% of the sample of
the intercoder reliability test as it has been accepted as the minimum percentage for a

random sample in order to have a good approximation of the population, as well for

37

allowing for enough cases to test for chance.164 An intercoder reliability of .80 and above
will be required for all variables.

A two-person coder reliability assessment produced a Scott's Pi that ranged ﬂ'om
.83 to 1.0. Speciﬁcally, Scott’s Pi computations for the variable of “Local Arab Comm.
Members” was .96; for “National Arab Comm. Members” .90; for “Local Government
Officials” 1.0; for “Federal Government Ofﬁcials” .91; for “Law Enforcement Officials”
1.0; for “Foreign Government Officials” 1.0; for “Geographical Focus” 1.0; for “Topical

Focus” .92; for “Stereotypes” .87; and for “Tone Towards Focus” .83.

Variables and Measurements

The hypotheses were tested using two dependent variables: news coverage and
stereotyping. The hypotheses had three independent variables: the deviance of Arab-
Americans and Islam as rated by the American public in a series of national surveys by
the Pew Foundation, the two types of newspapers (those with an Arab-American
community and those without one), and 9/11. Finally, the hypotheses had a control
variable: story origin. This last variable differentiates between staff-written stories and
stories generated ﬂ'om news agencies and control for wire copy as news agencies tend to
cover more conﬂict and breaking news stories than local community stories.165

The independent variables were selected in accordance with the premise of the
study, which is to investigate if there is a correlation between the deviant nature of a
group and the amount and type of news coverage it receives by newspapers. The Pew

surveys were selected as it is a reliable source of American public opinion about Islam

and Arab-Americans for a continuous period between 2001 and 2005. The Pew surveys

38

provided the necessary data that formed the independent variable, which is the deviance
of Arab-Americans and of Islam as rated by the American public. The dependent
variables were also chosen in accordance with the premise of study, with both of them —
reﬂected in the newspaper content - being investigated for signs of causation with the
independent variable.

“Deviance” was measured on three statements which are thought to be measure
political and religious deviance. The surveys by the Pew Forum on Religion & Public
Life were conducted in November 2001, March 2002, July 2003, July 2004, and July
2005. The measures of deviance are: 1) Islam’s Similarity to the religion of Most
Americans, 2) F avorability of Muslims and Muslim Americans, and 3) Islam’s
inclination toward violence. These categories were selected because they reﬂect the
views of the American public on two key areas that are part of study’s premise, and that
is how deviant is Islam as a religion and how deviant are Arab-Americans according to
the American public. These three categories encompass precisely those two key areas.
Studies on deviance have shown a connection between deviance and religion when
religion encourages deviant behavior,'66 which in this case is the view of the American
public about Islam. In extension of this perception of religion is deviant is the category
about how dissimilar is Islam from the religion of most Americans, as well as about the
unfavorable view of the American public about Arab-Americans who in their opinion are
Muslims.I67

l._1_slpm’s simiLagty to the remn of most Amerim: The survey question used

was: “From what you know, do you think that the Muslim religion and your own religion

have a lot in common, or do you think that the Muslim religion and your religion are very

39

different?” The question was included in the Pew surveys conducted in 2001 through
2005. In 2001 some 31% of the Americans surveyed said that Islam was similar to their
own, while 52% said that are very different. In 2005, the percentage of those Americans
who said that Islam was similar to their own religion dropped to 27% while the
percentage of those who said that Islam was very different ﬂom their own religion rose
to 59%.

2. Milky of Muslim-Americm: The survey question used was: “Is your
overall opinion of Muslim-Americans very favorable, mostly favorable, mostly
unfavorable, very unfavorable?” The question was included in the Pew surveys
conducted in 2001 through 2005. In 2001 some 59% of the Americans surveyed said they
had a favorable view about Muslim-Americans, while 17% said that they had unfavorable
views about Muslim-Americans. In 2005, the percentage of those Americans who said
that they had favorable views about Muslim-Americans dropped to 55%, while the
percentage of those who had unfavorable views about Muslim-Americans rose to 25%.

3._Isl_am’s inclinption towar:d violence: The survey question used was ﬂamed in
two statements and the interviewee had to choose which one came closer to their own
views even if neither is exactly right: “The Islamic religion is more likely than others to
encourage violence among its believers” or “The Islamic religion does not encourage
violence more than others.” The question was included in the Pew surveys conducted in
2002 through 2005. In 2002, some 51% of the Americans surveyed said the Islam doesn’t
encourage violence, while 25% said that Islam is more likely to encourage violence. In

2005, the percentage of those Americans who said that Islam doesn’t encourage violence

40

dropped to 47%, while the percentage of those who said that Islam is more likely to
encourage violence rose to 36%.

A content analysis was used to test the dependent variables of the hypotheses and
the variables for the research questions. The following variables where selected in
accordance with the premise of the study, and Speciﬁcally in addressing the four
hypotheses and the four research questions.

“News coveme” was measured by the proportion of hard news, features, and in-
depth stories published by the newspaper. This was achieved by counting the total
number of stories published in each newspaper and dividing by nurhber of days. This
variable was used to answer H1 and H4 and the unit of analysis was the story.

“M” was measured by a total of six variables: local Arab community
members; national Arab community members; local government ofﬁcials; federal
government ofﬁcials; law enforcement officials; and foreign government officials. The
six “sources” variables were selected because they allowed the study to identify who
were the most ﬂequently used sources in the stories about Arab-Americans, as well as to
see how much voice local Arab community members and national Arab community
members had in the stories dealing with their communities compared to omcial
government entities at the local, federal, and international level. A higher presence of
local Arab community members in the stories would be an indicator that the newspapers
made an effort to include local Arab views in the stories that were related about them or
issues about them, while a higher use of government sources would indicate that the
newspapers relied more heavily on official sources and not paying enough attention to the

views of the local Arab community. Each variable was compared with the other ﬁve

41

based on the ﬂequency it appears in the stories. This variable was used to answer RQ3
and the unit of analysis was the attributed statement within a story.

“M” was measured by two variables, Geographical Focus and Topical Focus.
Geographical Focus has three dimensions: a) local community story, b) localization of
national aﬂ‘airs, and c) localization of international affairs. The speciﬁc three categories
of the “geographical focus” variable were selected because they allowed the study to
identify if the stories about Arab-Americans were dealing with issues deriving from the
local Arab-American community or at the national and international levels. The second
variable, Topical Focus, has also three dimensions: a) local Arab commtmity affairs, b)
discrimination/attacks against Arabs, and 0) religion. The speciﬁc three categories of the
“topical focus” were selected because they allowed the study to identify if some key
topics were reﬂected in the coverage and how much. This variable was used to answer
RQ2 and the unit of analysis was the news story.

“Hard News, Feature Stories and In-depth Stories” was measured by a single
variable “type of story” that has three categories: hard news story; feature story; and in-
depth story. This variable sought to identify how many hard news, feature, and in-depth
stories where published by the eight dailies about Arab-Americans in comparison to other
types of content. It is argued that feature stories are much lengthier and detailed in their
analysis in comparison to hard news — published a newspaper. Features also require a lot
more time and resources than hard news, an indication that the news medium is showing
a stronger interest in an event or issue. In-depth stories differ ﬂom hard news and feature
stories. For the purposes of the study they include news analysis and investigative stories.

In-depth stories differ from hard news stories in that they go beyond pure facts about a

42

timely event and offer expert analysis of what the event means for present time and the
future. They also differ from features because they are not human interest stories. “In-
dep ” stories offer analysis and projection about current events through analysis from
experts on the topic. This variable was used to answer KO 1 and the unit of analysis was
the single hard news story, feature, and in-depth story.

“Stemming” was measured by a single variable — Stereotypes -— which has two
dimensions, presence or absence. Coders were instructed to observe for presence of eight
of stereotypes: 1) backward, 2) devious, 3) dirty, 4) dishonest, 5) corrupt, 6) violent, 7)
fanatic terrorists, and 8) anti-American. These stereotypes were selected from the
extensive scholarship on stereotyping and on portrayals of racial and ethnic minorities in
news media, and particularly of Arabs and Arab-Americans, as well as based on the
questions ﬂom the Pew national surveys. '68 This variable was used to answer H2 and H3
and the unit of analysis was a single word within each news story.

”Mp” will be measured by one variable: “Tone toward Focus.” The variable has
four categories: positive toward focus; balanced toward focus; negative toward focus; and
story is entirely neutral. This last variable — used in the “2006 News Availability Project”
as part of a national study for the Project for Excellence in Journalism — was selected
because it will allow the study to identify if the newspaper coverage of Arab-Americans
was positive, balanced, negative, or neutral, and if there was a change in the coverage
tone after 9/11. This variable will be used to answer KO 4 and the unit ofanalysis is a
single paragraph.

“Stag Orig ” was measured by one variable: “Story Origin.” The variable has

ﬁve categoriers: staff-written story; wire-written story; combo (staff-wire) story; other;

43

and no source. Staff-written stories are identiﬁed by either a byline, where the title of
“staff writer” or “Detroit News Writer” appears under the name of the author, or by a
simple identiﬁer as “staff report” or “The Detroit News” that appears at the top or bottom
of the story. Similarly, stories written by wire services are identiﬁed by a byline, where
the title of “Associated Press Writer” appears under the name of the author, or by a
simple identiﬁer as “The Associated Press” or “AP” that appears at the t0p or bottom of
the story. Stories that combine sources use simple identiﬁers such as “from Staff and

Wire reports” or “(The Detroit News/AP)” either at the top or the bottom of the story).

Data Analysis
The study used three different statistical tests to analyze data ﬂom the four

different samples. Speciﬁcally, for the “Full Census Sample,” which was used to address
H4, the analysis involved only the counting of news stories and comparing the total story
volume published by each newspaper during that period. The analysis tested for Crarner’s
V as a level of association. This test was deemed appropriate as the sample was a census,
meaning that any differences that were found in the sample exist in the population.
Crarner’s V test measures the strength of association -— or dependency — between two
nominal variables.

For the “Partial Census Sample,” which was used to answer H1 and HZ, two
different methods of analysis were used. Data ﬂom leas analyzed by comparing the
percentage of stories published by each newspaper in the three months following each
national survey by the Pew between 2001 and 2005. Data ﬂom H2 was analyzed by

comparing the total number of stories during that speciﬁc period that included

44

stereotypes. The data analysis included means and a Spearman’s correlation test and a
Crarner’s V as a level of association. The speciﬁc tests were chosen because it is as a
non-parametric measure of correlation can assess how well an arbitrary monotonic
ﬂmction can describe the relationship between two variables (“deviance” of Islam and
Arab—Americans and “news coverage” or “stereotyping”) while not making any
assumptions about the ﬂequency distribution of the variables. Spearman’s correlation test
reﬂects the degree of linear relationship between two variables, while Cramer’s V test
measures the strength of association — or dependency — between two nominal variables.

. Spearman’s correlation varies ﬂorn -l (a perfect negative conelation) to +1 (a
perfect positive correlation).169 The .20 to .40 range usually indicates a low positive
correlation — deﬁnite but small relationship. The .40 to .70 range usually indicates a
moderate positive correlation and substantial relationship. Finally, the .70 to 1.0 usually
indicates a strong positive correlation and an important relationship.170 Cramer’s V
ranges ﬂom 0 to +1, with 0 indicating no relationship and +1 indicating a perfect
relationship.171 A Cramer’s v of .2 or higher will be interpreted as being strong enough to
suggest a relationship.

For the “Main Random Sample,” which was used to address H3, the nominal data
were tested for difference of proportions, while Chi-square test of signiﬁcance and
Cramer’s V test for level of association were used to determine if differences in story
distributions found in the sample exist in the population. Chi-square test is used to
examine if distributions of categorical (nominal) variables differ ﬂ'om one another, while
Cramer’s V test measures the strength of association — or dependency — between two

nominal variables.

45

Finally, for the “Large Random Sample,” which was used to answer all four
research questions, the nominal data were tested for difference of proportions, while Chi-
square and Cramer’s V statistical tests were used to determine if differences in story
distributions found in the sample exist in the population. The two tests were deem
appropriate because Chi-square test is used to examine if distributions of categorical
(nominal) variables differ ﬂom one another, while Cramer’s V test measures the strength
of association — or dependency — between two nominal variables.

Chi-square is primarily a test of statistical signiﬁcance, rather than a measure of
strength of an association, because its value can range from very small to large depending
upon the size of the table (the number of rows and columns) and the size of the sample.'72
Also, Chi-square test requires that the expected frequencies in each cell not to be too
small -— deﬁantly higher than zero. In general, for a two-by-two contingency table (two
rows and two columns), the total number of cases should be greater than forty, and all
expected sell ﬂequencies should be ﬁve or more.‘73 In these cases, Chi-square cannot be

calculated. In the cases were Chi-square cannot be properly calculated, the variations in

the percentages of cases in cells will be discussed.

RESULTS

H1 states that the more deviant Arab-Americans are according to the American
Public, the more likely they are to be covered in the news by the Los Angeles Times, the
San Diego Union-Tribune, the Detroit Free Press, the Detroit News, the Oregonian, the
Milwaukee Journal Sentinel, the Denver Post, and the Minneapolis Star Tribune. The

hypothesis was tested by using the “Partial Census Sample.”

46

The ﬁndings (See Table 1) show that the hypothesis was only moderately
supported. The ﬁndings show that while US. public perceptions about the deviance of
Arab-Americans (in this case indicate by the unfavorable view of the US. public towards
then) increased by some 15% between 2001 and 2004, the coverage of Arab-Americans
by the newspapers that have vibrant Arab-American communities (Michigan and
California) showed a 5% decline in the coverage between 2002 and 2003 and otherwise
marginally increases during the other time periods. The newspapers with no, or non-
signiﬁcant, Arab-Americans communities in their area (Colorado, Minnesota, Oregon,
and Wisconsin) also Showed marginal declines in their coverage, a 4% drop between
2001 and 2002 and another 4% decline between 2003 and 2004 and otherwise marginally
increases during the other time periods.

The Shoemaker and Reese theory suggests a positive relationship between the
society’s ideologically perceived deviance of Arab-Americans and of the amount of
coverage that the group received by the media. '74 That is the more deviant Arab-
Americans are perceived by the US. public the more inclined the news media will be to
cover them. As shown in Table 1 that relationship is not always positive and a
Spearman’s Correlation produced a p = .35 — a low correlation between the perceived
deviance of Arab-Americans by the US. public (in this case the “unfavorable” view of
the US. public towards Arab—Americans) and the amount of coverage they received by
the media.

The ﬁndings suggest that locality (or marketplace) — that is the environment

within a news medium operates, identiﬁed by Shoemaker and Reese as part of the

47

extramedia inﬂuences — does play a role and perhaps an intervening factor in this case

against the ideological inﬂuences that derive from the greater society.

TABLE 1 Public Opinion about Deviance of Arab-Americans and
News Coverage by Location of Newspapers

 

 

Public Opinion Newspaper Coverage
Islam Percentage of Percentage of
Year Dissimilar Unfavorable Islam’s Stories Published Stories Published
to own of Muslims/ Inclination . by dailies b}: Dailies
Religion Arabs to Violence With 13186 Arab- Wlth small
American Arab-American
Communities Communities
2001 52% 17% -* 72% 28%
(76)*** (29)
2002 57% 29% 25% 76% 24%
(42) (13)
2003 60% 3 1% 44% 7 1% 29%
(27) (1 1)
2004 —* * 32% 46% 75% 25%
(27) (9)
2005 59% 25% 36% 71% 29%
(12) (5)
Total
N=241 174 67

 

* Question was not asked in that year’s survey.
** Question was not asked in that year’s survey.

**"‘ Number of stories about Arab-Americans published during that period.

48

Even when the newspapers are examined separately, the ﬁndings Show that there
is no clearly consistent pattern in the coverage. In the case of the Minneapolis Star
Tribune the only time period that there is an important increase in coverage is between
2004 and 2005, when its coverage increased ﬂom .03% to 23%. During that period,
though, the unfavorable view of the US. public about Arab-Americans declined by some
7%, ﬂom 32% to 25%. The ﬁndings also show inconsistencies in the coverage even
among dailies that share Similarities to locality. For example, the Michigan newspapers
did not follow a similar pattern in their increases and decreases of coverage of Arab-
Americans during the periods investigated, nor did the Californian newspapers. All four
of those publications cater to the two largest Arab—American communities in the nation.
The ﬁndings also show inconsistencies in the coverage among the four dailies that do not
cater to localities that have any signiﬁcant Arab-American communities.

H2 states that the more deviant Arab-Americans or events involving Arab-
Americans are according to the American public the more likely they are to be
stereotyped in the news coverage of the Los Angeles Times, the San Diego Union-
Tribune, the Detroit Free Press, the Detroit News, the Oregonian, the Milwaukee Journal
Sentinel, the Denver Post, and the Minneapolis Star Tribune. The hypothesis was tested
by using the “Partial Census Sample.”

The ﬁndings (See Table 2) do not provide support to the hypothesis. Speciﬁcally,
the four dailies with large Arab-American communities in the three-month period of 2001
had 5% (or four stories) out of their 76 stories with stereotypes. That ﬁgure in 2002 and
2003 dropped to 0%, only to increase in 2004 to 15% (or four stories out of 27) and to

drop again to 8% (or one story out of 12) in 2005. A Cramer’s V of .322 suggests that

49

there is a weak but notable relationship between type of community and presence of
stereotyping. Still, there is no clear positive relationship between the public’s perceived
deviance of Arab-Americans and stereotyping in the news coverage. Similarly, the four
dailies with small Arab-American communities in 2001 had 21% (or six stories) out of 29
stories with stereotypes. That ﬁgure in 2002 dropped to 0%, only to rise in 2003 to 9%
(or one story out of 11) and to drop again in 2004 and 2005 to 0%. Again, the trends in
this speciﬁc type of dailies do not show a clear positive relationship between the public’s
perceived deviance of Arab-Americans and stereotyping in the news coverage.

The percentage of stories with stereotypes about Arab-Americans that were
published by newspapers with the two largest Arab-American communities in the nation
(Michigan and California) ranged ﬂour 0% to 15%. The percentage of stories with
stereotypes about Arab-Americans that were published by newspapers with small
presences of Arab-American communities (Colorado, Minnesota, Oregon, and
Wisconsin) also ranged from 0% to 21%. Moreover, the examination of the content
showed that the presence of stereotypes in those stories was due to the nature of the
stories and not generated by the newspapers. In other words, it was not the news media
that was using stereotypes against Arab-Americans, but it was because they were
stereotyped by others that it became news. The newspapers were Simply reporting on the
stereotypes and attributing them to sources in the stories.

The ﬁndings run contrary to the Shoemaker and Reese proposition that news
media stereotype minority groups deemed as deviant by society. Shoemaker’s research

found that the news media do not screen out deviant ideas but rather portray them in a

50

way calculated to underscore their deviance.175 The ﬁndings did not support this positive
relationship.

Similarly with H1, it is unclear what intervening factors may have contributed to
the variance of the ﬁndings. Locality, that is the environment within a newspaper
operates, does not appear to be a factor in this hypothesis as both types of newspapers —
those with large Arab-American communities and those with small Arab-American
communities — had periods with no stereotypes in their stories and periods with increases

and decreases in stereotypes.

51

TABLE 2 Public Opinion about the Deviance of Arab-Americans and Presence of
Stereotypes in News Coverage by Location of Newspapers

 

 

Public Opinion Newspaper Coverage
Percentage of Percentage of
Year Islam Unfavorable Islam’s Stereotypes in Stereotypes in
Dissimilar of Muslims/ Inclination Stories Published Stories Published
to own Arabs to Violence by (1311133 by Dailies
Religion with large Arab- with small
American Arab-American
Communities Commmlities
2001 52% 17% -* .05% 21%
(4/76)*** (6/29)
2002 57% 29% 25% 0% 0%
(0) (0)
2003 60% 3 1% 44% 0% 0.9%
(0) (1/1 1)
2004 -** 32% 46% 15% 0%
(4/27) (0)
2005 59% 25% 36% .08% 0%
(1/12) (0)
Total
N=241 174 67

 

* Question was not asked in that year’s survey.

** Question was not asked in that year’s survey.

*** Number of stories that were published about Arab-Americans and included
stereotypes.

Cramer’s V=.322

52

Even when examining the ﬁndings based on case-by-case of each newspaper the
ﬁndings do not Show any clear pattern, Two newspapers, the Denver Post and the
Minneapolis Star Tribune had no stories that included stereotypes in any of the ﬁve
periods that were investigated. The newspaper with the highest percentage of stories with
stereotypes was the Detroit Free Press with 33%.

H3 states that the Los Angeles Times, the San Diego Union-Tribune, the Detroit
Free Press, and the Detroit News will be less likely to include stereotypes about Arab-
Americans in their coverage with the Oregonian, the Milwaukee Journal Sentinel, the
Denver Post, and the Minneapolis Star Tribune. The hypothesis was tested using the
“Main Random Sample” and its N=500 covers the entire 1996-2006 period.

The ﬁndings (See Tables 3 & 4) do not support the hypothesis. Speciﬁcally, the
ﬁndings Show that there was only a 2% difference between the newspapers that have
large Arab-American communities in their area (Michigan and California) and those that
have non-signiﬁcant Arab-American communities in their area (Colorado, Minnesota,
Oregon, and Wisconsin) when it came to stories that included stereotypes about Arab-
Americans. Those with large Arab-American communities had 15% (or 56 stories) of
their stories with stereotypes while those with non-signiﬁcant Arab-American
communities had 13% (or 16 stories) of their stories with stereotypes. A Cramer’s V of
.234 suggests a weak relationship between type of community and percentage of stories
with Arab-American stereotypes.

The Shoemaker and Reese proposition on “extramedia inﬂuences” argues that the
more a media organization promotes itself within a target audience, the more its content

will reﬂect the interests of that audience. ”6 This hypothesis explored this assumption that

53

the extramedia inﬂuences are stronger than ideological inﬂuences but the ﬁndings did not
support it.

Moreover, the examination of the content showed that the presence of stereotypes
in those stories was due to the nature of the stories and not generated by the newspapers.
In other words, it was not the news media that was using stereotypes against Arab-

Americans, but it was because they were stereotyped by others that it became news.

TABLE 3 Percentage of Stories with Stereotypes in Coverage of Arab-Americans
by Region from 1996 to 2006

 

Years Percentage of Stories with Percentage of Stories with
Stereotypes in Coverage by Dailies Stereotypes in Coverage by Dailies
with large Arab—American with small Arab-American
Communities Communities

1996 — 15% 13%

2006 (56)* (16)

Total

N = 500 374 126

 

* Number of stories published about Arab-Americans that included stereotypes

Cramer’s V=.234, p-.000

54

TABLE 4 Percentage of Stories with Stereotypes in Coverage of Arab—Americans
By Each Newspaper from 1996 to 2006

 

Newspapers

 

Year LA San Oregonian Detroit Detroit Milwaukee Denver Minneap

Times Diego Free News 1.8. Post olis
U—T Press Star
Tribune

1996- .08% .02% 28% 25% 14% .06% l 1% 12%

2006 (7)* (1) , (7) (32) (16) (3) (3) (3)
Total 85 45 25 130 1 14 50 26 25
N=500

 

* Number of stories that were published about Arab-Americans and included stereotypes

Cramer’s V=.234, p=.000

Even when examining ﬁndings based on case-by-case of each newspaper the
ﬁndings do not show any clear correlation. The two Michigan newspapers had more
stories with stereotypes than any other newspaper, except the Oregonian that had seven
stories, or 28%, with stereotypes. Speciﬁcally, of the 130 stories of the Detroit Free
Press that were examined, 25% (or 32 stories) included stereotypes, while of the 114
stories of the Detroit News that were examined, 14% (or 16 stories) included stereotypes.
This runs contrary to the Shoemaker and Reese proposition of “extramedia inﬂuences” on
media content. Examination of those stories, though, showed that the presence of
stereotypes in those stories was due to the nature of the stories and not generated by the
newspapers. In other words, it was not the news media that was using stereotypes against

Arab-Americans, but it was because they were stereotyped by others that it became news.

55

The newspapers were simply reporting on the stereotypes and attributing them to sources
in the stories.

H4 states that the news coverage of Arab-Americans by the Los Angeles Times,
the San Diego Union-Tribune, the Detroit Free Press, the Detroit News, the Oregonian,
the Milwaukee Journal Sentinel, the Denver Post, and the Minneapolis Star Tribune will
increase after the terrorist attacks of 9/11. The hypothesis was tested using the “Full
Census Sample.”

The ﬁndings (See Table 5) support the hypothesis only partially. Speciﬁcally, the
ﬁndings show that the four dailies with large Arab-American communities had a constant
increase in the amount of coverage of Arab-Americans between 1996 and 2001, but then,
the amount started to decrease every year until 2006 when it rose again. A similar pattern
of coverage was also found in the case of the four dailies with small Arab-American
communities, with coverage of Arab-Americans increasing almost constantly between
1996 and 2001 — except 1999 when it decreased by 11% - while after 2001 it started to
decrease every year until the end of 2006.

Even when examining each newspaper separately, the ﬁndings show that six of
the eight newspapers (the Los Angeles Times, the Detroit News, the Detroit Free Press,
the San Diego Union Tribune, and the Milwaukee Journal Sentinel) increased their
coverage of Arab-Americans between 1996 and 2006, and two newspapers (the
Oregonian and the Denver Post) decreased their coverage after a temporary increase.
Still, even those that increased their coverage of Arab-Americans between 1996 and 2006
had some decreases during the time period. For example, the Detroit Free Press had 18

stories about Arab-Americans in 1996 and steadily increased its coverage to reach 117

56

stories in 2001. But, following 2001 the coverage started to decrease: 2002, 74 stories;
2003, 55 stories; 2004, 46 stories; and 2005, 16 stories. In 2006 the Detroit Free Press
increased its coverage once again to 61 stories. Similar were the patterns of all six
newspapers that saw an increase of their coverage of Arab-Americans between 1996 and
2006.

The Shoemaker and Reese theory argues that the more deviant an event is the
more coverage it will generate.'77 In this case the terrorist attacks of Sept. 11, 2001
served as the case of the deviant event. The ﬁndings show that in 2001 there was an
increase in the coverage of Arab-Americans across all eight newspapers, supporting the
hypothesis. But, ﬁve of the eight newspapers (the San Diego Union Tribune, the
Oregonian, the Detroit Free Press, the Detroit News, and the Denver Post) saw a
decrease in their coverage of Arab-Americans as early as the following year. Thus, the
ﬁndings show that although there was a certainly an increase in the coverage of Arab-
Americans following the deviant event of 9/11, that increase of coverage was not

sustainable.

57

TABLE 5 Total Number of Stories about Arab-Americans Published by Region
From 1996 To 2006

 

Total Number of Stories Total Number of Stories
Year Published by Dailies with Published by Dailies with
large Arab-American Small Arab-American
Communities Communities
34* 1 l
1996 (.09%) (13%)
35* 15
1997 (10%) (16%)
53* 27
1998 (1 5%) (29%)
104 17
1999 (29%) (18%)
134 22
2000 (37%) (24%)
Total (1996-2000) N=360 N=92
309 120
2001 (26%) (31%)
258 98
2002 (22%) (25%)
235 91
2003 (20%) (24%)
143 36
2004 (12%) (.09%)
73 25
2005 (06%) (.07%)
153 17
2006 (14%) (.04%)
Total (2001-2006) N=1,l71 N=387
Total N=2,010 1,531 479

 

* Online archival database for Detroit Free Press did not years prior to 1999.

RQl asks about the number of feature and in-depth stories that the Los Angeles
Times, the San Diego Union-Tribune, the Detroit Free Press, the Detroit News, the
Oregonian, the Milwaukee Journal Sentinel, the Denver Post, and the Minneapolis Star
Tribune carried about Arab—Americans pre- and post-9/11. The research question was
tested by using the “Large Random Sample.”

The ﬁndings (See Tables 6, 7 & 8) show that the number of feature stories
published by the four dailies with large Arab-American communities increased by 6% in
the post-9/11 period, while the four dailies with small Arab-American communities saw
an increase of 5%. At the same time, in-depth stories published by the four dailies with
large Arab-American communities saw a .02% rise in the post-9/11 period, while the
same type of stories in the four dailies with small Arab-American communities had a 5%
decrease. Hard news accounted for the majority of the stories (60% or more) of all eight
dailies in both time periods. In both time periods the two Michigan newspapers carried
more than 50% (21 stories in the pre-9/11 period and 60 stories in the post-9/11 period)
of the number of feature stories that were published by the eight dailies. A Cramer’s V of
more than .24 in both time periods suggests that there is a marginal relationship between

the type of community and the type of stories published by a newspaper.

59

TABLE 6 Type of Stories about Arab-Americans Published
by Region Five Years Prior and Five Years after 9/11

 

Type of Stories Published by
Dailies with large Arab-American

Type of Stories Published by Dailies
with small Arab-American

Communities Communities
Type of Pre-9/ 1 1 Post-9/1 1 Pre—9/ l 1 Post-9/ l 1
Story
Feature 30 78 7 32
. (23%) (29%) (20%) (25%)
In- 10 24 7 19
depth (.07%) (.09%) (20%) (15%)
Hard 89 165 20 75
News (70%) (62%) (60%) (60%)
Total
N=554 127 267 34 126

 

Pre-9/11 Sample: X2 (14, N=161) =21.97, p=.079, Cramer’s V=.261, p=.079

Post-9/ll Sample: X2 (14, N=393) =46.58, p=.000, Cramer’s V=.243, p=.000

60

TABLE 7 Type of Stories about Arab-Americans Published by Each Newspaper
Five Years Prior to 9/11

 

 

Newspapers
Type LA San Orego Detroit Detroit Milwau Denver Minneap
of Times Diego nian Free News kee Post olis
Story U-T Press Journal Star
Sentinel Tribune
Feature 7 0 2 10 1 l 1 2 2
(21%) (0%) (20%) (1 8%) (3 7%) (10%) (22%) (40%)
In- 5 2 3 2 1 3 1 0
depth (1 5%) (20%) (30%) (.04) (.03%) (30%) (1 1%) (0%)
Hard 21 8 5 42 18 6 6 3
News (64%) (80%) (50%) (78%) (60%) (60%) (67%) (60%)
Total
N=16l 33 10 10 54 30 10 9 5

 

X2 (14, N=161) =21.97, p=.079, Cramer’s V=.26l, p=.079

61

TABLE 8 Type of Stories about Arab-Americans Published by Bach Newspaper

 

 

Five Years after 9/11
Newspapers
Type LA San Orego Detroit Detroit Milwau Denver Minnea-
of Times Diego nian Free News kee Post polis
Story U-T Press Journal Star
Sentinel Tribune
Fea- 10 8 7 22 3 8 5 1 1 9

ture (19%) (18%) (28%) (26%) (45%) (10%) (42%) (36%)

In- 9 8 0 4 3 9 5 5
depth (17%) (18%) (0%) (.05%) (.03%) (18%) (19%) (20%)

Hard 34 29 18 53 44 36 10 11
News (64%) (64%) (72%) (69%) (52%) (72%) (39%) (44%)

Total
N=393 53 45 25 84 85 50 26 25

 

Cramer’s V=.243, p=.000

RQ2 asks about the focus of the Arab-related stories of the Los Angeles Times, the
San Diego Union-Tribune, the Detroit Free Press, the Detroit News, the Oregonian, the
Milwaukee Journal Sentinel, the Denver Post, and the Minneapolis Star Tribune pre- and
post-9/1 1. The research question was tested by using the “Large Random Sample.”

Story focus involved two variables, Geographical Focus and Topical Focus of a
news story. Both variables required that in order for a story to be coded in a certain
category (e. g. local community affairs) it should cover more than 50% of the entire story.
In terms of the Geographical Focus of stories the ﬁndings (See Table 9) show

that local community stories published by the four dailies with large Arab-American

62

communities saw a 2% percent decline in the post-9/11 period, ﬁom 55% (or 70 stories)
to 53% (or 142 stories). At the same time, the same type of stories saw an 18% rise in the
four dailies with small Arab-American communities. At the same time, stories involving
localized national affairs that were published by the four dailies with large Arab-
American communities saw a marginal decline of 0.2%, ﬁom .09% (or 11 stories) to
.07% (or 19 stories) in the post-9/11 period, while in the case of the four dailies with
small Arab-American communities it remained steady at .09%. Finally, stories involving
localization of international affairs that were published by the four dailies with large
Arab-American communities saw a decline of 3% in the post-9/1 1 period, from 12% (or
15 stories) to .09% (or 25 stories). In the case of the four dailies with small Arab-
American communities, the trend for that type of stories was identical, showing a 3%
decline ﬁom 15% (or ﬁve stories) to 12% (or 15 stories). A Cramer’s V of about .3
indicates that there is a weak but notable relationship between the type of community and
the geographical focus of the stories published by a newspaper.

In examining each newspaper separately, local community stories accounted for
48% (or 77 and 190 stories respectively) published by the eight dailies in the pre- and
post-9/11 period. The Michigan newspapers again accounted for the bulk of those stories
(69%, or 53 stories, in the pre-9/11 period and 61%, or 117 stories, in the post-9/11
period). Localization of international affairs stories accounted for 12% (20 stories in the
pre-9/1l period and 40 stories in the post-9/11 period) published by the eight dailies,
while localization of national affairs accounted for less than 1% (14 stories in the pre-

9/11 period and 31 stories in the post-9/ll period).

63

TABLE 9 Geographical Focus of Stories about Arab-Americans
Published by Region Five Years Prior and Five Years after 9/11

 

Geographical Focus of Stories Geographical Focus of Stories
Published by Dailies with large Published by Dailies with small
Arab-American Communities Arab-American Communities
Geographical Pre-9/l l Post-9/ 1 1 Pre-9/ 1 1 Post-9/ l 1
Focus
Local 70 142 7 48
Community (55%) (53%) (20%) (38%)
Story
Localization
of National 11 19 3 12
Affairs (.09%) (.O8%) (.09%) (10%)
iLocalization {
f of Inter/nal l 5 25 5 15
. Affairs (12%) (.09%) (1 5%) (12%)
Other/ Not 3 1 81 1 9 5 1
Applicable (24%) (30%) (56%) (40%)
Total
N=554 127 267 34 126

 

Pre-9/11 Sample: X2 (28, N=161) =64.98, p=.000, Cramer’s V=.318, p=.000

Post-9/ll Sample: X2 (28, N=393) =2.14, p=.000, Cramer’s V=.369, p=.000

64

In terms of the Topical Focus of stories the ﬁndings (See Table 10) show that
stories about Local Arab Community A ﬂairs that were published by the four dailies that
have large Arab-American communities saw a 3% decline, ﬁom 42% (or 53 stories) to
3% (or 104 stories). At the same time, this type of stories in the four dailies with small
Arab-American communities saw an 18% increase, from zero to 18% (or 23 stories).
Stories about Discrimination & Attacks against Arabs that were published by the four
dailies with large Arab-American communities remained steady at 14% in the post-9/11
period, while the same type of stories in the four dailies with small Arab-American
communities saw a 4% increase, from 15% (or ﬁve stories) to 19% (or 24 stories). Stories
about Religion that were published by the four dailies with large Arab-American
communities remained steady at .02% in the post-9/11 period, and similarly the same
type of stories in the four dailies with small Arab-American communities remained
steady at .03%. Again, a Cramer’s V of about .3 suggests that there is a weak but notable
relationship between the type of community and the topical focus of the stories published
by a newspaper.

In looking at each newspaper separately, stories about Local Arab Community
Affairs accounted for 33% (or 53 stories) of the stories published by the eight dailies in
the pre-9/11 period and for 32% (or 127 stories) in the post-9/11 period. Again, the two
Michigan dailies accounted for the bulk of those stories (90%, or 48 stories, in the pre-
9/11 period and 72%, or 92 stories, in the post-9/11 period). Discrimination & Attacks
against Arabs accounted for 20% (or 23 stories) in the pre-9/11 period and for 15% (or
61 stories) in the post-9/11 period. The Los Angeles Times accounted for 43% (or 10

stories) of those stories in the pre-9/11 period, while the Detroit Free Press accounted for

65

23% (or 14 stories) of those stories. Finally, Religion accounted for less than 1% in both

periods (4 stories in the pre-9/ 11 period and 10 stories in the post-9/11 period.

TABLE 10 Topical Focus of Stories about Arab-Americans
Published by Region Five Years Prior and Five Years after 9/11

 

Topical Focus of Stories Topical Focus of Stories
Published by Dailies with large Published by Dailies with small
Arab-American Communities Arab-American Communities
Topical Focus Pre-9/ l l Post-9/1 l Pre-9/1 ll Post—9/ l 1
Local Arab 53 104 0 23
Community (42%) (39%) (0%) (18%)
Affairs
Discrimination
Attacks 18 37 5 24
Against Arabs (14%) (14%) (15%) (19%)
Religion 3 6 1 4
(.02%) (.02%) (.03%) (.03%)
Other/ Not 53 120 28 75
Applicable (42%) (45%) (82%) (60%)
Total
N=554 127 267 34 126

 

Pre-9/11 Sample: Cramer’s V=.332, p=.000

Post-9/11 Sample: Cramer’s V=.308, p=.000

66

RQ3 asks about the types of sources that appeared most often in the stories about
Arab-Americans or Arab affairs published in the Los Angeles Times, the San Diego
Union-Tribune, the Detroit Free Press, the Detroit News, the Oregonian, the Milwaukee
Journal Sentinel, the Denver Post, and the Minneapolis Star Tribune pre- and post-9/11.
The research question was tested by using the “Large Random Sample.”

The ﬁndings (See Table 11) show that there was a decline in Local Arab
Community Sources in the post-9/11 coverage of both types of newspapers (those with
large Arab-American communities and those with small Arab-American communities),
while at the same time there was a large increase in Law Enforcement Sources.
Speciﬁcally, stories citing Local Arab Community Sources that were published by the
four dailies with large Arab-American populations saw a 4% decline in the post-9/11
period, from 70% (or 89 stories) to 66% (or 177 stories). At the same time, stories citing
the same type of sources that were published by the four dailies with small Arab-
American communities saw a 2% decline, from 38% (or 13 stories) to 36% (or 45
stories). Meanwhile, stories citing Law Enforcement Sources that were published by the
four dailies that were published by the four dailies with large Arab American
communities saw a 10% increase in the post-9/11 period, from 11% (or 14 stories) to
21% (or 57 stories). At the same time, stories citing the same type of sources that were
published by the four dailies with small Arab-American communities saw a 30% increase
in the post-9/11 period, from .06% (or two stories) to 36% (or 45 stories). Stories citing
National Arab Community Members remained steady in both periods for the four dailies
with large Arab-American communities, while a 10% increase was recorded for the four

dailies with small Arab-American communities, from 35% (or 12 stories) to 45% (or 57

67

stories). The ﬁndings also show a 4% increase in the post-9/11 use of Local Government
Oﬁ‘icial sources by four dailies with small Arab-American communities, from .03% (or
one story) to .07% (or nine stories). In the case of pre-9/11 period Cramer’s V is above
.2, suggesting a weak relationship between the type of community and the type of sources
appearing in the stories published by a newspaper, but in the case of post-9/ 11 period
Cramer’s V falls below .2, suggesting the relationship is not very strong between the type
of community and the type of sources appearing in the stories published by a newspaper.

In examining each newspaper separately, the ﬁndings show that Local Arab
Community Sources accounted for 63% (or 102 stories) in pre-9/11 stories published by
the eight dailies, and 56% (or 222 stories) in the post-9/11 period - a decline of 7%. At
the same time, Law Enforcement Sources saw a sharp increase. Stories that cited Law
Enforcement Sources accounted for 1% (or 16 stories) in the pre-9/11 period, and 26%
(or 102 stories) in the post-9/11 period.

The newspapers that had the most stories that cited Local Arab Community
Sources were again the Michigan dailies, accounting for 71% (or 71 stories) out of 102
that cited Local Arab Community Sources in the pre-9/11 period and for 65% (or 144
stories) out of 222 that cited Local Arab Community Sources in the post 9/11 period. The
newspaper that had the most stories that cited Law Enforcement Sources in the post-9/ 11
was the Milwaukee Journal Sentinel, accounting for 22% (or 23 stories) out of the 102

that cited Law Enforcement Sources.

68

TABLE 11 Type of Sources in Stories about Arab-Americans
Published by Region Five Years Prior and Five Years after 9/11

 

Type of Sources in Stories
Published by Dailies with large
Arab-American Communities

Type of Sources in Stories
Published by Dailies with small
Arab-American Communities

Types of Pre-9/ 1 1 Post-9/ l 1 Pre-9/ l 1 Post-9/1 1
Source“
Local Arab 89 177 13 45
Community (70%) (66%) (3 8%) (3 6%)
Members
National
Arab 39 82 12 57
Community (31%) (31%) (35%) (45%)
Members
Local
Government 14 34 1 9
Officials (l 1%) (13%) (.03%) (07%)
Federal
Government - 39 77 14 46
Ofﬁcials (3 1%) (29%) (41%) (3 6%)
Law
Enforcement 14 57 2 45
Officials (11%) (21%) (.06%) (36%)
Foreign
Government 10 16 5 7
Ofﬁcials (08%) (.06%) (15%) (.05%)
Total
N=554 127 267 34 126

 

* One or more of the same source per story count as one unit.
Pre-9/1l sample: X2 (35, N=l61) =54.72, p=.018, Cramer’s V=.261, p=.018

Post-9/11 sample: X2 (28, N=393) =49.19, p=.008, Cramer’s V=.177, p=.008

69

RQ4 asks about the tone toward the focus of the Arab-related stories published by
the Los Angeles Times, the San Diego Union-Tribune, the Detroit Free Press, the Detroit
News, the Oregonian, the Milwaukee Journal Sentinel, the Denver Post, and the
Minneapolis Star Tribune pre- and post-9/11. The research question was tested by using
the “Large Random Sample.”

Tone toward focus was determined by counting the number of negative, positive
and neutral paragraphs of the story that were related to its main focus. In order for a story
to be coded positive or negative, the number of paragraphs (positive or negative) should
amount to more than 66% of the story’s main focus.

The ﬁndings (See Table 12) show that the tone of the coverage of Arab
Americans by both types of newspapers (those with large Arab—American communities
and those with small Arab-American communities) was mostly balanced during both
periods. Speciﬁcally, the four dailies with large Arab-American communities saw a 1%
increase in balanced stories in the post-9/11 period, from 54% (or 69 stories) to 55% (or
147 stories). At the same time, the same type of stories published by the four dailies with
small Arab-American communities saw a 16% decline, from 65% (or 22 stories) to 49%
(or 62 stories). In terms of stories with positive tone, in the case of the four dailies with
large Arab-American communities there was a 2% increase in the post-9/11 period, from
16% (or 20 stories) to 18% (or 47 stories), while in the case of the four dailies with small
Arab-American communities there was a 6% increase, from .03% (or one story) to .09%
(or 11 stories). In terms of the stories with negative tone, the four dailies with large Arab-
American communities there was a 3% decline in the post-9/11 period, ﬁom 27% (or 34

stories) to 24% (or 65 stories), while in the case of the four dailies with small Arab-

7O

American communities, there was a 14% increase, ﬁ‘om 23% (or eight stories) to 37% (or
47 stories). In terms of entirely neutral stories, in the case of the four dailies with large
Arab-American communities there was no change between the two periods, remaining
steady at .03%, while in the case of the four dailies with small Arab-American
communities there was a 2% increase, from .03% (or one story) to .05% (or six stories).
In this case, Cramer’s V is above .2during both time periods, suggesting a weak
relationship between the type of community and the tone of the stories published by a
newspaper.

In examining each newspaper separately, the ﬁndings show that the eight dailies
did not have any signiﬁcant changes ﬁ'orn the ﬁve years prior to 9/11 to the ﬁve years
after the terrorist attacks. Speciﬁcally, stories with positive tone accounted for 13% (or
21 stories out of 161) of the stories published by the eight dailies in the pre-9/11 period,
and for 15% (or 58 stories out of 393 stories) in the post-9/ll period — a 2% increase.
Stories with balanced tone accounted for 56% (or 91 stories out of 161) of the stories
published by the eight dailies in the pre-9/ 11 period, and for 53% (or 209 out of 393) in
the post-9/11 period —- a 3% decline. Stories with negative tone accounted for 26% (or 42
stories out of 161) of the stories published by the eight dailies in the pre-9/11 period, and
for 28% (112 stories out of 393) in the post-9/11 period — a 2% increase. Finally, stories
that were entirely neutral in tone accounted for .04% (or 7 stories out of 161) of the
stories published by the eight dailies in the pre-9/1 period, and for .03% (or 14 stories out
of 393) in the post-9/11 period — a .01% change.

Thus, the coverage not only remained the same in terms of tone — changes ranged

between .01% and 3% - but also in its majority was balanced (56% pre-9/11 and 53%

71

post-9/11). The dailies with the most stories with positive tone in the pre- and post-9/ 1 1
periods was The Detroit News, accounting for 43% (or nine stories out of 21) of the
stories that had positive tone and were published by the eight dailies in the pre-9/11
period, and for 41% (or 24 stories out of 58) in the post-9/11 period. At the same time,
the daily that had the most stories with negative tone in the post-9/ 11 period was the
Detroit Free Press, accounting for 33% (or 14 stories out of 42) of the stories that had
negative tone and were published by the eight dailies during that period. Finally, the daily
with the most stories with negative tone in the post-9/1 1 period was the Milwaukee
Journal Sentinel, accounting for 22% (or 25 stories out of 112) of the stories that had

negative tone and were published by the eight dailies during that period.

72

TABLE 12 Tone toward Focus of Stories about Arab-Americans
Published by Region Five Years Prior and Five Years aﬁer 9/11

 

Tone in Stories Published by Tone in Stories Published
Dailies with large Arab- by Dailies with small
American Communities Arab-American Communities
Tone Pre-9/1 1 Post-9/1 l Pre-9/ 1 1 Post-9/1 1
Toward
Focus
Positive 20 47 1 11
(16%) (18%) (.03%) (.09%)
Balanced 69 147 22 62
(54%) (55%) (65%) (49%)
Negative 34 65 8 47
(27%) (24%) (23%) (3 7%)
Entirely 4 8 3 6
Neutral (.03%) (.03%) (.09%) (.05%)
Total
N=554 127 267 34 126

 

Pre-9/ll sample: X2 (21, N=l61) =27.42, p=.157, Cramer’s V=.223, p=.195

Post-9/11 sample: X2 (21., N=393) =73.49, p=.000, Cramer’s V=.250, p=.00

73

DISCUSSION

Overall, the results support the notion that deviance plays some role in the news
coverage of Arab-Americans, but the role is not as strong as the theory would predict.
The ﬁndings show that deviance is correlated with some types of coverage but not with
others. The ﬁndings indicate that there may be at least three possible reasons for the lack
of correlation in some instances: 1) the impact of deviance varies with topic, 2) the
impact varies with individual newspapers because editors and reporters play an important
role as gatekeepers in screening perceptions of deviance about events or groups, and 3)
the relative inﬂuence of various factors that changed across time.

As such, the ﬁndings of the study failed to support or provided only partial and
moderate support for the Shoemaker and Reese Theories of Inﬂuence on Mass Media
Content, and particularly the inﬂuences of perceived deviance of a minority group by the
public on media content.

Contrary to the prediction by the theories of inﬂuence on mass media content by
Shoemaker and Reese, the study found that an increase in the US. public’s perceived
deviance of Arab-Americans did not result in increased stereotyping of them by eight
major metro newspapers in the ﬁve years that followed the deviant event of the Sept. 11,
2001 terrorist attacks.

Speciﬁcally, the ﬁndings from H2 show that the four dailies with large Arab-
American communities in the three-month period of 2001 had 5% (or four stories) out of
their 76 stories with stereotypes. That ﬁgure in 2002 and 2003 dropped to 0%, only to

increase in 2004 to 15% (or four stories out of 27) and to drop again to 8% (or one story

74

out of 12) in 2005. Still, there is no clear positive relationship between the public’s
perceived deviance of Arab.Americans and stereotyping in the news coverage. Similarly,
the four dailies with small Arab-American communities in 2001 had 21% (or six stories)
out of 29 stories with stereotypes. That ﬁgure in 2002 dropped to 0%, only to rise in 2003
to 9% (or one story out of 11) and to drop again in 2004 and 2005 to 0%. Again, the
trends in this speciﬁc type of dailies do not show a clear positive relationship between the
public’s perceived deviance of Arab-Americans and stereotyping in the news coverage.

The percentage of stories with stereotypes about Arab-Americans that were
published by newspapers with the two largest Arab—American communities in the nation
(Michigan and California) ranged from 0% to 15%. The percentage of stories with
stereotypes about Arab-Americans that were published by newspapers with small
presences of Arab-American communities (Colorado, Minnesota, Oregon, and
Wisconsin) also ranged from 0% to 21%. Moreover, the examination of the content
showed that the presence of stereotypes in those stories was due to the nature of the
stories and not generated by the newspapers. In other words, it was not the news media
that was using stereotypes against Arab-Americans, but it was because they were
stereotyped by others that it became news. The newspapers were simply reporting on the
stereotypes and attributing them to sources in the stories.

As suggested by the Shoemaker research, the news media do not screen out
deviant ideas but rather portray them in a way calculated to underscore their deviance. ' 78
In this case the ﬁndings do not support this argument. The ﬁndings indicate that other

factors are interplaying in this case and contributing to the variance of the ﬁndings. Still,

75

the current data do not allow for further analysis as to which speciﬁc factors are
intervening here.

The ﬁndings from H1 also provided only moderate support to the hypothesis that
the more deviant the US. public perceives Arab-Americans the more coverage they will
receive by the news media — also a hypothesis of the Shoemaker and Reese theory.

The ﬁndings show that while US public perceptions about the deviance of Arab-
Americans (in this case indicate by the unfavorable view of the US. public towards then)
increased by some 15% between 2001 and 2004, the coverage of Arab-Americans by the
newspapers that have vibrant Arab-American communities (Michigan and California)
showed a 5% decline in the coverage between 2002 and 2003 and otherwise marginally
increases during the other time periods. The newspapers with no, or small, Arab-
Americans communities in their area (Colorado, Minnesota, Oregon, and Wisconsin) also
showed marginal declines in their coverage, a 4% drop between 2001 and 2002 and
another 4% decline between 2003 and 2004 and otherwise marginally increases in the
other time periods.

The Shoemaker and Reese hypothesis indicates a positive relationship between
the society’s ideologically perceived deviance of Arab-Americans and of the amount of
coverage that the group received by the media.179 That is the more deviant Arab-
Americans are perceived by the US. public the more inclined the news media will be to
cover them. This study found a low correlation with a Spearman’s Correlation produced a
p = .35.

The ﬁndings indicate that deviance plays a role in affecting coverage, but other

factors play a role as well. Still, the current data do not allow for further analysis as to

76

which speciﬁc factors are intervening here and contributing to the variance of the
ﬁndings.

In investigating H4 the study also found that although there was an increase in
coverage of Arab-Americans after the terrorist attacks of 9/1 1 — as the Shoemaker and
Reese theory predicted — that increase was not present also in the following years. Thus,
the hypothesis that the news coverage of Arab-Americans by the Los Angeles Times, the
San Diego Union-Tribune, the Detroit Free Press, the Detroit News, the Oregonian, the
Milwaukee Journal Sentinel, the Denver Post, and the Minneapolis Star Tribune will
increase after the terrorist attacks of 9/11, an event that was deviant by nature, was only
partially supported.

Speciﬁcally, the ﬁndings show that there was only a 2% difference in the amount
of coverage between the newspapers that have large Arab-American communities
(Michigan and California) and those that have non-signiﬁcant Arab-American
communities (Colorado, Minnesota, Oregon, and Wisconsin) when it came to stories that
included stereotypes about Arab-Americans. Those with large Arab-American
communities had 15% (or 56 stories) of their stories with stereotypes while those with
non-signiﬁcant Arab-American communities had 13% (or 16 stories) of their stories with
stereotypes. Shoemaker and Reese proposition on “extramedia inﬂuences” argues that the
more a media organization promotes itself within a target audience, the more its content
will reﬂect the interests of that audience.180 This hypothesis explored this assumption that
the extramedia inﬂuences are stronger than ideological inﬂuences but the ﬁndings did not

support it.

77

Again, the ﬁndings indicate that other factors are interplaying in this case and
contributing to the variance of the ﬁndings. Still, the current data do not allow for further
analysis as to which speciﬁc factors are intervening here.

In investigating H3 the ﬁndings did not support the hypothesis that newspapers
that have large Arab-American communities in their area (the Los Angeles Times, the San
Diego Union-Tribune, the Detroit Free Press, and the Detroit News) will be less likely to
include stereotypes about Arab-Americans in their coverage than those with non-
signiﬁcant Arab-American communities in their area (the Oregonian, the Milwaukee
Journal Sentinel, the Denver Post, and the Minneapolis Star Tribune).

Speciﬁcally, the ﬁndings show that there was only a 2% difference between the
newspapers that have large Arab-American communities in their area (Michigan and
California) and those that have small Arab-American communities in their area
(Colorado, Minnesota, Oregon, and Wisconsin) when it came to stories that included
stereotypes about Arab-Americans. Those with large Arab-American communities had
15% (or 56 stories) of their stories with stereotypes while those with non-signiﬁcant
Arab-American communities had 13% (or 16 stories) of their stories with stereotypes.

The Shoemaker and Reese proposition on “extramedia inﬂuences” argues that the
more a media organization promotes itself within a target audience, the more its content
will reﬂect the interests of that audience. '81 This hypothesis explored this proposition that
the extramedia inﬂuences are stronger than ideological inﬂuences but the ﬁndings did not
support it.

Moreover, the examination of the content showed that the presence of stereotypes

in those stories was due to the nature of the stories and not generated by the newspapers.

78

In other words, it was not the news media that was using stereotypes against Arab-
Americans, but it was because they were stereotyped by others that it became news. The
newspapers simply reported the stereotypes and attributed them to sources in the stories.

Once again, the ﬁndings indicate that other factors are interplaying in this case
and contributing to the variance of the ﬁndings. Still, the current data do not allow for
further analysis as to which speciﬁc factors are intervening here.

The ﬁndings hour the research questions provide additional insights into the
speciﬁcs of the coverage. Perhaps one of the most interesting ﬁndings comes ﬁom RQ4
which asked about the tone of the coverage by the eight major metro dailies pre- and
post-9/ l 1.

The ﬁndings show that the tone of the coverage of Arab Americans by the eight
dailies did not have any large changes ﬁom the ﬁve years prior to 9/11 to the ﬁve years
after the terrorist attacks. The ﬁndings show that the tone of the coverage of Arab
Americans by both types of newspapers (those with large Arab-American communities
and those with small Arab-American commtmities) was mostly balanced during both
periods. Both types of newspapers also saw a nearly equal amount of increase in their
stories with positive tone in the post-9/11 period (between 2% and 6% respectively). In
terms of the stories with negative tone, the four dailies with large Arab-American
communities there was a 3% decline in the post-9/ll period, ﬁorn 27% (or 34 stories) to
24% (or 65 stories), while in the case of the four dailies with small Arab-American
communities, there was a 14% increase, from 23% (or eight stories) to 37% (or 47

stories).

79

Another ﬁnding was that the two Michigan dailies were not similar in the tone of
the coverage of Arab-American. The Detroit News was the daily with the most stories
with positive tone in the pre- and post-9/11 period, accounting for 43% (or nine stories
out of 21) of the stories that had positive tone and were published by the eight dailies in
the pre-9/11 period, and for 41% (or 24 stories out of 58) in the post-9/11 period. At the
same time, the daily that had the most stories with negative tone in the post-9/11 period
was the Detroit Free Press, accounting for 33% (or 14 stories out of 42) of the stories
that had negative tone and were published by the eight dailies during that period.

Finally, the daily with the most stories with negative tone in the post-9/ 11 period
was the Milwaukee Journal Sentinel, accounting for 22% (or 25 stories out of 112) of the
stories that had negative tone and were published by the eight dailies during that period.

Another ﬁnding from the data generated by the research questions was that both
types of newspapers (those with large Arab-American communities and those with small
Arab-American communities) saw a nearly identical amount of increase (between 6%
and 5% respectively) in the number of feature stories published in the post-9/11 period.
At the same time, in—depth stories published by the four dailies with large Arab-American
communities saw a .02% rise in the post-9/11 period, while the same type of stories in the
four dailies with small Arab-American communities had a 5% decrease. Hard news
accounted for the majority of the stories (60% or more) of all eight dailies in both time
periods.

In further examining the ROI by each newspaper separately, the ﬁndings show
that the Michigan dailies had more than 50% of all the feature stories published by the

eight dailies about Arab-Americans in the pre- and post-9/11 periods; had more than 60%

80

of all local community stories in the pre- and post-9/11 periods; had more than 70% of
local Arab community affairs stories in the pre- and post-9/11 periods; and had more than
60% of stories with one or more local Arab community sources in the pre- and post-9/11
periods.

The ﬁndings also showed that the number of feature stories published by the eight
newspapers increased by as much as 6% in the post-9/11 period (ﬁ'om 35 stories in the
pre-9/ll period to 110 in the post-9/11 period), while the number of in-depth stories
increased marginally, less than 1%. In both time periods the two Michigan newspapers
carried more than 50% (21 stories in the pre-9/11 period and 60 stories in the post-9/11
period) of the number of feature stories that were published by the eight dailies.

RQ2 asked about the focus of the Arab-related stories published by the eight
dailies pre- and post-9/11. In terms of the Geographical Focus of stories the ﬁndings
show that while the four dailies with large Arab—American communities saw a 2%
decline in local community stories in the post-9/11 period, the four dailies with small
Arab-American communities saw an 18% increase in the same type of stories. Stories
involving localized national affairs saw only marginal changes in the post-9/11 period by
both types of newspapers, while stories involving localization of international affairs saw
a 3% decline by both types of dailies in the post-9/11 period.

In terms of the Topical Focus of stories the ﬁndings show that stories about Local
Arab Community Aﬂairs published by the four dailies that have large Arab-American
communities saw a 3% decline in the post-9/11, while in the case of the four dailies with
small Arab-American communities the same type of stories saw an 18% increase. Stories

about Discrimination & Attacks against Arabs that were published by the four dailies

81

with large Arab-American communities remained steady at 14% in the post-9/ 11 period,
while the same type of stories in the four dailies with small Arab—American communities
saw a 4% increase, ﬁom 15% (or ﬁve stories) to 19% (or 24 stories). In terms of stories
about Religion both types of newspapers maintained their pre-9/ 11 levels of about 3%.

In examining each newspaper separately, in terms of the Geographical Focus of
stories the ﬁndings show that local community stories accounted for 48% (or 77 and 190
stories respectively) published by the eight dailies in the pre- and post-9/11 period. The
Michigan newspapers again accounted for the bulk of those stories (69%, or 53 stories, in
the pre-9/11 period and 61%, or 117 stories, in the post-9/1 1 period). Localization of
international affairs stories accounted for 12% (20 stories in the pre-9/11 period and 40
stories in the post-9/ll period) published by the eight dailies, while localization of
national affairs accounted for less than 1% (14 stories in the pre-9/11 period and 31
stories in the post-9/11 period).

In terms of the Topical Focus of stories the ﬁndings show that stories about Local
Arab Community Aﬂairs accounted for 33% (or 53 stories) of the stories published by the
eight dailies in the pre-9/11 period and for 32% (or 127 stories) in the post-9/11 period.
Again, the two Michigan dailies accounted for the bulk of those stories (90%, or 48
stories, in the pre-9/11 period and 72%, or 92 stories, in the post-9/ll period).
Discrimination & Attacks against Arabs accounted for 20% (or 23 stories) in the pre-9/11
period and for 15% (or 61 stories) in the post-9/11 period. The Los Angeles Times
accounted for 43% (or 10 stories) of those stories in the pre-9/11 period, while the Detroit

Free Press accounted for 23% (or 14 stories) of those stories. Finally, Religion accounted

82

for less than 1% in both periods (4 stories in the pre-9/ll period and 10 stories in the
post-9/11 period.

RQ3 asked about the types of sources that appeared most oﬁen in the stories about
Arab-Americans or Arab affairs published by the eight dailies pre- and post-9/11. The
ﬁndings show that there was a decline in Local Arab Community Sources in the post-9/1 1
coverage of both types of newspapers (those with large Arab-American communities and
those with small Arab-American communities), while at the same time there was a large
increase in Law Enforcement Sources.

Speciﬁcally, stories citing Local Arab Community Sources that were published by
the four dailies with large Arab-American populations saw a 4% decline in the post-9/11
period, from 70% (or 89 stories) to 66% (or 177 stories). At the same time, stories citing
the same type of sources that were published by the four dailies with small Arab-
American communities saw a 2% decline, ﬁom 38% (or 13 stories) to 36% (or 45
stories).

Meanwhile, stories citing Law Enforcement Sources that were published by the
four dailies that were published by the four dailies with large Arab American
communities saw a 10% increase in the post-9/11 period, from 11% (or 14 stories) to
21% (or 57 stories). At the same time, stories citing the same type of sources that were
published by the four dailies with small Arab-American communities saw a 30% increase
in the post-9/11 period, from .06% (or two stories) to 36% (or 45 stories).

The newspapers that had the most stories that cited Local Arab Community
Sources were again the Michigan dailies, accounting for 71% (or 71 stories) out of 102

that cited Local Arab Community Sources in the pre-9/11 period and for 65% (or 144

83

stories) out of 222 that cited Local Arab Community Sources in the post 9/11 period. The
newspaper that had the most stories that cited Law Enforcement sources in the post-9/11
was the Milwaukee Journal Sentinel, accounting for 22% (or 23 stories) out of the 102
that cited Law Enforcement Sources.

Once again, the ﬁndings indicate that other factors are interplaying in this case in
inﬂuencing the news media content. Still, the current data do not allow for ﬁirther
analysis as to which speciﬁc factors are intervening here and contributing to the variance
of the ﬁndings.

Finally, an analysis of story origin, who wrote a story, in relation to stereotypes
and tone showed that staff-generated coverage of Arab-Americans had a higher
percentage of stories with stereotypes (17%, or 70 stories out of 41 3) than those ﬁ'om
wire services 4%, or three stories out of 79). When it came to tone, though, the ﬁndings
were reversed, with the wire-generated coverage having a percentage of stories with
negative tone (44%, or 35 stories out of 79) than those written by staffers at the eight
dailies (25%, or 104 stories out of 413). Staff-generated coverage also had a higher
percentage of stories with positive tone (16%, or 66 stories out of 413) than the wire-
generated coverage (9%, or seven stories out of 79). These ﬁndings suggest that the
inﬂuences on news content at the ideological level affected the two types of writers
differently. This is inconsistent with the Shoemaker and Reese model, which says
ideology should affect all journalists.

As with many studies, this one has limitations. A larger sample of dailies could
have generated more data and insights but constraints in time and resources made a larger

sample restrictive. There are also limitations in terms of generalizations, as this study

84

focused only in print media. Future studies can build on these ﬁndings by investigating
the coverage of Arab-Americans by electronic media during the same time period and by
examining an even larger sample of daily newspapers. In doing so, future studies can
continue expanding our knowledge and understanding of the relationship between the

perceived deviance of minority groups by society and the coverage of those groups.
CONCLUSIONS

As shown by the ﬁndings, the study provides only moderate support to the
Shoemaker and Reese Theories of Inﬂuence on Mass Media Content, and particularly the
inﬂuences of perceived deviance of a minority group by the public on media content.

Results showed that the more deviant the U.S. public perceives Arab-Americans
the more coverage they will receive by the news media. However, contrary to the
prediction by the theories of inﬂuence on mass media content, the study found that an
increase in the U.S. public’s perceived deviance of Arab-Americans was not correlated
with increased stereotyping of them by eight major metro newspapers in the ﬁve years
that followed the Sept. 11, 2001 terrorist attacks.

Understanding the contradictory results requires a look at the theory. The
Shoemaker and Reese theory contends that the ideological layer of inﬂuence is the most
powerful one as all the processes taking place at the lower layers are thought to be
working toward an ideologically related pattern of messages and on behalf of the higher
power centers of society.182 Under this model, society is said to have a direct ideological

inﬂuence on news media content, with journalists and editors becoming aware of

85

society’s beliefs through mediated (e.g. surveys) or interpersonal (e.g. face-to-face)
communication, and adopting and transferring those views in the news content. This
model contends that the individual media worker has the least inﬂuence on media
content.

The Pew surveys clearly show that the U.S. public’s perceived deviance of Arab—
Americans was increasing steadily between 2001 and 2005. The fact that news coverage
by the eight dailies did not result in any signiﬁcant presence of stereotyping of Arab-
Americans during that period shows that other factors were at play in the process and in a
way disrupted the ﬂow of inﬂuences on content. Speciﬁcally, there is an array of
processes that take place from the time a journalist or editor is informed of society’s
views and the time that content is produced:

0 One of the main considerations that journalists and editors have to take into
consideration is ethics. Each news outlet has its own code of ethics that reﬂect the
general industry-wide code of ethics. For example, the U.S.-based Society of
Professional Journalists notes in its code of ethics that journalists should “avoid
stereotyping by race, gender, age, religion, ethnicity, geography, sexual
orientation, disability, physical appearance or social status.”l83 Similarly, the
Associated Press code of ethics notes that its journalists should “not use
Obscenities, racial epithets or other offensive slurs in stories unless they are part of

direct quotations and there is a compelling reason for them.”184
0 Another signiﬁcant consideration that journalists and editors have to take into

account is their market (or audience). For instance, journalists and editors

working at a newspaper that is published in an area that has a vibrant Jewish

86

community will be inclined to take into consideration the sensitivities of their
audience on ethnic, political and religious issues that are relevant to their market
before publishing a story about it.

Societal-level inﬂuences are also “ﬁltered” through the individual media workers.
Breed (1955) found ﬁve cases where staffers (or individual media workers)
manage to bypass content policy that was instituted by the media organization. "85
Breed’s ﬁndings are equally relevant today as reporters continue to take on
specialized beats that allow them to develop in-depth knowledge about speciﬁc
topics, while some reporters continue to support the “star system,” where their
elevated status over other colleagues gives them greater autonomy over content-
related decisions.

These processes take place at an individual level (by each journalist and editor)

and at a collective level (editorial meetings). These processes are not mutually exclusive,

meaning that a journalist can either adopt or reject those societal-level inﬂuences based

on their professional code of ethics, while they can also discuss those inﬂuences at a daily

editorial meeting where the editorial staff decides what will be published.

Based on the Pew survey results the amount of coverage about Arab-Americans

should have increased annually aﬁer the terrorist attacks of 9/11, but the ﬁndings showed

that it only increased in 2001 and then followed a downward trajectory.

The fact that news coverage of Arab-Americans by the eight dailies did not follow

a upward trend after 9/11 indicates that other factors were at play in the process.

Speciﬁcally, there are several reasons that could explain as to why the coverage of Arab-

Americans did not continue to increase after 2001:

87

o Downsizing of newsrooms. In recent years newspapers across the country have
been struggling ﬁnancially, and as a result they have been cutting down the
number of their employees. According to the American Society of News Editors
census, 2,400 full-time professional newsroom jobs were lost at American dailies
in 2007 and 5,900 more in 2008.186 That amounts to daily newspapers losing
about 17% of their news staffs since the start of 2001 .187 This means fewer
journalists to cover news, thus less news coverage.

0 Decrease in newspaper size and ﬁequency of publication. Higher newsprint prices

(up about 25%) caused newspapers to cut back space devoted to news. '88

Moreover, several papers already, and likely more in 2009, are taking the half-
step of deleting the traditional paper several days a week to save on production
and delivery costs, with the hope that it will not cost them much in advertising or
alienate their print audiences. This means less space in the newspaper for stories
and thus perhaps a change in coverage priorities.

0 News values and speciﬁcally the timeliness events. After the events of 9/11 and
the subsequent events related to those terrorist attacks, coverage of Arab-
Americans was losing its timeliness because of a decline in major events
surrounding them. As the timeliness of events involving Arab-American started to
decline, other timely events started to take their position in the news coverage.

Any of these factors, individually or collective, could have played a role in the

decline of coverage about Arab-Americans ﬂowing 2001, indicating that previously

assumed constants in news coverage are challenged by changes in the news media

industry.

88

In conclusion, it is suggested that the Shoemaker and Reese’s theory was only
moderately supported by the ﬁndings of this study in part because of factors that were
developed in the years following the publication of their theory. Industry-wide ethics
have become more important as newspapers seek to establish or maintain their credibility
among their audiences; newspapers are struggling ﬁnancially and that has resulted in
downsizing of newsrooms, fewer pages in a newspaper, and even fewer print days during
a week. These changes are powerful enough to cause changes to the processes by which
news media produce their content.

The research on which the theory was based took place over four decades. It
seems likely that some of the relationships found in this research may have changed as
journalism reacted to its environment.

Perhaps the Shoemaker and Reese theoretical model needs to be revisited based
on the existing conditions in industry and society and newer research on news media,
allowing it to be more aligned with current realities. Future research could revisit the
theory by exploring how changes in the industry affect news production processes, and
thus the theoretical model that was published by Shoemaker and Reese nearly 20 years
ago. With newspapers moving from the traditional print format to an online-only format,
space for stories will no longer be an issue; cost-reduction from the print production
could allow for the hiring of more newsroom staff; and newsroom dynamics and ethics
may be subject to change. Perhaps the nature of ideology and its inﬂuence have changed
as well. These are issues that future research could explore and perhaps lead to an

evolution of the Shoemaker and Reese theoretical model.

89

APPENDIX

Content Analysis Protocol

V01 — STORY SLUG

DEFINITION: Write up to three words that will give this story a unique sum slug.
V02 - ID NUMBER

DEFINITION: Each story receives an ID number starting with 0001.

V03 — NEWSPAPER

DEFINITION: This variable identiﬁes the news mediaLoutlet from which stories were

taken. It designates, in 1 digit, the media genre and outlet.

Los Angeles Times = 1
San Diego Union-Tribune = 2
Oregonian = 3
Detroit Free Press = 4
Detroit News = 5
Milwaukee Journal Sentinel = 6
Denver Post = 7
Minneapolis Star Tribune = 8

V04 — NEWS SOURCE

DEFINITION: This variable identiﬁes the source of the news story. Stories published in
newspapers can come from their own staff, wire services, or can combine sources (i.e.
wire and staff). Staff-written stories are identiﬁed by either a byline, where the title of

“staff writer” or “Detroit News Writer” appears under the name of the author, or by a

90

simple identiﬁer as “staff report” or “The Detroit News” that appears at the top or bottom
of the story. Similarly, stories written by wire services are identiﬁed by a byline, where
the title of “Associated Press Writer” appears under the name of the author, or by a
simple identiﬁer as “The Associated Press” or “AP” that appears at the top or bottom of
the story. Stories that combine sources use simple identiﬁers such as “from Staff and
Wire reports” or “(The Detroit News/AP)” either at the top or the bottom of the story). If

the source of the story is other, then code 4. If the source of the story is unidentiﬁable,

then code 0.

Staff-Written Story = 1
Wire-Written Story = 2
Combo (Staff-Wire) Story = 3
Other = 4
No source = 0

V05 — TYPE OF STORY

DEFINITION: This variable indicates, in 1 digit, the type of story, which includes hard
news, and features.

“Hard news” involve stories that report on a speciﬁc timely event or series of events, with
a focus on what, when, where, and why, and which commonly appear on the hunt page
or any other pages in the news section.

“EM” these are also known as human-interest stories because they usually focus on
prominent or unusual events and individuals. They differ from hard news stories in that
the topic is not necessarily timely and that they are much lengthier. In newspapers this

type of stories usually appear in a “Features” section, but some newspapers don’t have a

91

dedicated section for them. In some cases feature stories appear within specialized
sections, such as “Local” or “State.” Investigative stories are also included in “Features.”
“News Analysis” differs from hard news and feature stories. They differ from hard news
stories in that they go beyond pure facts about a timely event and offer expert analysis of
what the event means for present time and the future. They also differ ﬁ'om features
because they are not human interest stories. “News Analysis” stories offer analysis and
projection about current events through analysis from experts on the topic. They are

commonly identiﬁed by a “News Analysis” logo either next to the headline or inside the

story.

Hard News Story = 1
Features Story = 2
News Analysis = 3

IMPORTANT NOTE: The following six (6) variables — V5 though V10 — are story
source variables and each one of them is concerned with a different type of sources.
Please read carefully the description of each source variable before you start the coding
for sources.

V06 — LOCAL ARAB COMMUNITY MEMBERS

DEFINITION: Local Arab Community members are any persons who are identiﬁed in
the article as being of Arab descent and who reside in the Arab community in the city,
county, or state where the newspaper is published. Common identiﬁers include “Arab-

American,” “Arab community member,” “Arab community leader,” “local Imam,” “local

92

Arab religious leader,” etc. Arab Community members include but are not limited

ordinary citizens, businessmen, religious and community leaders, and any advocates.

One Local Arab Comm. Member = 1
Two Local Arab Comm. Members = 2
Three Local Arab Comm. Members = 3
Four of more local Arab Comm. Members = 4
No Local Arab Comm. Member = 0

V07 — NATIONAL ARAB COMMUNITY MEMBERS

DEFINITION: National Arab Community members are any persons who are indentiﬁed
in the story as being of Arab descent and who reside in an Arab community outside the
state where the newspaper is published. Common identiﬁers include “Arab-American,”
“Arab community member,” etc. National Arab Community members include but are not

limited to ordinary citizens, businessmen, religious and community leaders, and any

advocates.

One National Arab Comm. Member = 1
Two National Arab Comm. Members = 2
Three National Arab Comm. Members = 3
Four of more National Arab Comm. Members = 4
No National Arab Comm. Member = 0

93

V08 - LOCAL GOVERNMENT OFFICIALS
DEFINITION: Local Government Ofﬁcials include any city, county, and state ofﬁcials
from the area where the newspaper is published. These include but are not limited to

mayors, city council members, county commissioners, county clerk, and governors. Law

enforcement officials are NOT included here.

One Local Government Ofﬁcial = 1
Two Local Government Officials = 2
Three Local Government Ofﬁcials = 3
Four or more Local Government Ofﬁcials = 4
No Local Government Ofﬁcials = 0

V09 - FEDERAL GOVERNMENT OFFICIALS

DEFINITION: Federal Government Ofﬁcials include any elected or appointed official of
the federal government. These include but are not limited to the president of the United
States, any members of the administration (such as any spokesperson of various
departments and the White House, and the secretaries of the departments, including
Homeland Security), members of Congress and Senate, and any ofﬁcial of the federal

government. Law enforcement officials are NOT included here.

One Federal Government Official = 1
Two Federal Government Officials = 2
Three Federal Government Ofﬁcials = 3
Four of more Federal Government Officials = 4
No Federal Government Ofﬁcials = 0

94

V10 — LAW ENFORCEMENT OFFICIALS

DEFINITION: Law Enforcement Officials include any city, county, state, or federal law
enforcement officials. These include but are not limited to state police, sheriffs, and law
enforcement agents of federal agencies such as the FBI and the Naturalization and
Immigration Service. Former law enforcement officials who comment on a case they
were involved in the past are also included here. This includes former law enforcement

officials who comment on current events related to their work.

One Law Enforcement Official = 1
Two Law Enforcement Officials . = 2
Three Law Enforcement Ofﬁcials = 3
Four of more Law Enforcement Ofﬁcials = 4
No Law Enforcement Ofﬁcials = 0

V1] - FOREIGN GOVERNMENT OFFICIALS
DEFINITION: Foreign Government Ofﬁcials include any elected or appointed civilian or
military official of a foreign government. These include but are not limited to presidents,

prime ministers, ministers, military and law enforcement officials.

One Foreign Government Official = 1
Two Foreign Government Officials = 2
Three Foreign Government Officials = 3
Four or more Foreign Government Officials . = 4
No Foreign Government Officials = 0

95

IMPORTANT NOTE: The next two (2) variables — V11 through V12 — are story focus
variables and each one of them is concerned with different dimensions of story focus.

Please read all variable deﬁnitions carefully before starting the coding of story focus.

V12 — GEOGRAPHICAL FOCUS

DEFINITION: Stories can focus on three different entities: a) Local Community Story, b)
Localization of National Affairs, and c) Localization of International Affairs.

In order to be coded here, the type of the focus must be more than 51% (type of focus/
story length) of the story. The percentage will be calculated based on the total number of
paragraphs in the story (e.g. if the story has 10 paragraphs, then more than ﬁve (5) of
them will have to be dealing with one of the three focuses). If the main focus of the story
is other than the ones identiﬁed here then code ﬁve (5). If the story doesn’t have a single
focuses reaching or exceeding the 51% then code nine (9).The types of focus are deﬁned
as follow:

Local Communig Story: This occurs when the majority of a story concentrates on an
event or issue that is generated within the state where the newspaper is published. This
type of story does not preclude local national or international affairs from being
mentioned, but the emphasis is on the local affair.

Localization of National Affairs: This occurs when the majority of a story concentrates
on a national event or issue, such as the passing of a bill by Congress, an act by the
President or the administration and how the local Arab community reacts to those
national events or issues. This type of story does not preclude local community issues or

international affairs ﬁ'om being mentioned, but the emphasis is on the national affair.

96

L_oc_ilzation of International Affairs: This occurs when the majority of a story
concentrates on an international event or issue, such as the conﬂict in the Middle East
between Israelis and Palestinians or the Iraq War and how the local Arab community
reacts to those events or issues. This type of story does not preclude local community

issues or national affairs ﬁ'om being mentioned, but the emphasis is on the international

affair.

Local Community Story = 1
Localization of National Affairs = 2
Localization of International Affairs = 3
Other = 5
Not applicable = 9

V13 - TOPICAL FOCUS

DEFINITION: Stories can focus on three different entities: a) Local Arab community
affairs, b) Discrimination/Attacks against Arabs, and c) Islamic Religion. In order to be
coded here, the type of the focus must be more than 51% (type of focus/ story length) of
the story. The percentage will be calculated based on the total number of paragraphs in
the story (e. g. if the story has 10 paragraphs, then more than ﬁve (5) of them will have to
be dealing with one of the three focuses). If the main focus of the story is other than the
ones identiﬁed here then code ﬁve (5). 1f the story doesn’t have a single focuses reaching

or exceeding the 51% then code nine (9). The types of focus are deﬁned as follow:

97

Local Arab Community Affairs: This occurs when the majority of a story concentrates on
an event or issue that is generated within the local Arab-American community or within
the state where the newspaper is published and has a direct impact on the local Arab-
American community. This does not preclude discrimination/attacks against Arabs,
national or religion being mentioned, but the emphasis is on an event or an issue that is
generated ﬁorn within the local Arab-American community.

Discrimination/Attacks against Arabs: This occurs when the majority of a story
concentrates on discrimination or attacks against Arab-Americans locally or nationally.
This does not preclude local Arab community affairs or religion ﬁ'orn being mentioned,
but the emphasis is on incidents of discrimination or attacks against Arab-Americans.
This also includes stereotyping.

Islamic Relig'on: This occurs when the majority of a story concentrates on the Islamic
religion, including issues about the Islamic religion, Islamic religious practices, and
Islamic religious events. This type of story can be either local or national and it does not
preclude local Arab community affairs or discrimination/attacks against Arabs ﬁ'om

being mentioned, but the emphasis is on the Islamic religion.

Local Arab Community Affairs = 1
Discrimination/Attacks against Arabs = 2
Religion = 3
Other = 5
Not applicable = 9

98

V14 - STEREOTYPES

DEFINITION: The variable measures the presence or absence of negative stereotypes
about Arab-Americans. Coders should be examining each story closely for presence of
the following stereotypes that are attributed to Arab-Americans: 1) backward, 2) devious,
3) dirty, 4) dishonest, 5) corrupt, 6) violent, 7) fanatic terrorists, and 8) anti-American. It
is IMPORTANT that the stereotypes are attributed directly to Arab-Americans (e. g. “the
anti-American sentiment among Arab communities in the United States is growing”). If
in doubt, then code 0. If one of the stereotypes mentioned above is present in a story then
code 1. If none of these stereotypes is present in a story then code 0.

Stereotypes absent = 0

Stereotypes present = 1

V15 — TONE TOWARD FOCUS

DEFINITION: The variable measures the positive, neutral or negative tone toward the
focus of the story as identiﬁed in V13. Quotes and Assertions can come from the sources
in the story, while assertions can also come from the journalist.

Steps:

1. Coders determine whether win) is positive. neutral. or negative
covegge toward the story’s focus. Types of characterization include, but are not
conﬁned to: approve/disapprove; admire/criticize; acknowledge success/point out
failure; exceed/fall short of expectations; veracity vindicated/contradicted;
portrayed as object of respect/ridicule; optimism/pessimism for the future. For

these characterizations to be considered, they must address the focus of the story.

99

2. Some stories may be entirely neutral in reporting. If no negative or positive
, statements are found in the story, code 9.

3. Coders then count the percentages of all positive and negative paragraphs and
code the story according to the 2:] Rule.
The 2:1 Rule: This rule requires that “tone” coding decisions be reached by
quantifying the positive and negative quotes, assertions, interpretations and
innuendos made by the journalist or “others” within the news story; To be coded
as positive or negative, the ratio must reach or exceed 2:1 (2/3) or (66%) in either

direction. If not, 2/3, it is balanced.

Positive toward focus =1
Balanced toward focus =2
Negative toward focus =3
Story is entirely neutral =9

100

ENDNOTES

 

1 Colin Summer, The Sociology of Deviance: An Obituary. (New York:
Continuum, 1994); Emile Durkheim, The Rules of Sociological Method. (London,
Macmillan, 1982 ,frrst published 1893).

2 Pamela J. Shoemaker and Stephen D. Reese, Mediating the Message; Theories
of Inﬂuences on Mass Media Content. (White Plains, NY: Longman Publishers,
1 996).

3 Pamela J. Shoemaker, “The Perceived Legitimacy of Deviant Political Groups:
Two Experiments on Media Effects,” Communication Research, 9 (1982): 249-
286; Douglas M. McLeod, “Communicating Deviance: The Effects of Television
News Coverage of Social Protests,” Journal of Broadcasting & Electronic Media,
39 (1) (winter 1995): 4-20.

4 Alex Heckert and Druann M. Heckert, “Using an Integrated Typology of
Deviance to Analyze Ten Common Norms of the U.S. Middle Class,”
Sociological Quarterly, 45 (2) (spring 2004): 209-228.

5 Shoemaker and Reese, Mediating the Message.

6 Serge Moscovici, “Toward a Theory of Conversion Behavior,” in L. Berkowitz
(ed.), Advances in Experimental Social Psychology, vol. 13 (New York:
Academic Press, 1980).

7 Bevery A. D. Keever, Carolyn Martindale, and Mary Ann Weston, U.S. News
Coverage of Racial Minorities: A Sourcebook, 1 934-1996. (W estport, CT:
Greenwood Press, 1997).

8 Keever, Martindale, and Weston, U.S. News Coverage.
9 Shoemaker and Reese, Mediating the Message.

10 Joan Brodsky Schur, Coming to America: The Arabs. (F armington Mills:
Greenhaven

Press, 2005); The Arab-American Institute. “Demographics.” Retrieved from:
http://www.aaiusa.org/arab-americans/22/demographics (March 30, 2007); U.S.
Census 2000 Special Reports: “We the People of Arab Ancestry in the United
States,” March 2005. Retrieved ﬁ'om:
http://www.census.gov/prod/2005pubs/censr-21 .pdf (March 30, 2007); U.S.
Census: “The Arab Population: 2000,” December 2003. Retrieved from:
http://www.census.gov/prod/2003pubs/c2kbr-23.pdf (March 30, 2007).

101

 

11 Michael Welch, Scapegoats of September 1 1‘”: Hate Crimes & State Crimes in
the War on Terror. (New Jersey: Rutgers University Press, 2006).

‘2 Shoemaker and Reese, Mediating the Message.
‘3 Shoemaker and Reese, Mediating the Message.

14 University of Oregon. “Hutchins’ Commission Report.” Retrieved from:
http://jcomm.uoregon.edu/~cbybee/j20l/overheads/hutchins_sensational.htm
(Nov. 16, 2008).

15 University of Oregon. “Hutchins’ Commission Recommendations.” Retrieved
from: http:// j comm.uoregon.edu/~cbybee/ j20 l /overheads/hutchins_code.htm
(Nov. 16, 2008).

'6 Keever, Martindale, and Weston, U.S. News Coverage.

17 Pamela J. Shoemaker, Tsan-Kuo Change and Nancy Brendlinger, “Deviance as
a Predictor of Newsworthiness: Coverage of International Events in U.S. Media,”
in Communication Yearbook 10, ed. M.L. McLaughlin (Newbury Park, CA: Sage,
1987); Shoemaker, “The Perceived Legitimacy of Deviant Political Groups”;
McLeod, “Communicating Deviance”; Shoemaker, “Media Treatment of Deviant
Political Groups,” Journalism Quarterly, 61 (spring 1984); 66-75; Pamela J.
Shoemaker, Lucig Danielian and Nancy Brendlinger, “Deviant Acts, Risky
Business, and U.S. Interests: The Newsworthiness of World Events,” Journalism
Quarterly, 69 (winter 1991): 781-95; Pamela Shoemaker, “The communication of
deviance,” in B. Dervin, ed., Progress in Communication Science (Norwood, NJ;
Ablex, 1987): 172.

18 Shoemaker and Reese, Mediating the Message.
'9 The Arab-American Institute. “Demographics.”

2° Christianity Today (2003). “Are Most Arab Americans Christian?” Retrieved
ﬁom:

http://www.christianitytoday.com/ct/2003/marchweb-only/3-24-22.0.html (Non.
l, 2008); The Brookings Institute. “Arab and Muslim America: A Snapshot.”
Retrieved from:
http://www.brookings.edu/articles/2002/winter_middleeast_telhami.aspx (Nov. 1,
2008); American-Arab Anti Discrimination Committee. “Facts About Islam.”
Retrieved from: http://www.adc.org/index.php?id=249 (Nov. 3, 2008); Encarta
MSN. “Arab Americans.” Retrieved from:
http://encarta.msn.com/encyclopedia_76 l 587469/arab_amcricans.html (Nov. 1,
2008).

102

 

2‘ American-Arab Anti Discrimination Committee. “Facts About Islam”; The
Detroit Free Press (2001). “100 Questions and Answers about Arab-Americans: A
Joumalist’s Guide.” Retrieved from:
http://www.freep.com/jobspage/arabs/index.htm (Nov. 3, 2008).

22 The Detroit Free Press. “100 Questions and Answers about Arab-Americans.”

23 Pew Forum on Religion & Public Life (2002). “Americans Struggle With
Religion’s

Role at Home and Abroad,” March 2002. Retrieved from:
http://peoplepress.org/reports/display.php3?PageID=3 87 (March 20, 2007).

24 Pew Forum on Religion & Public Life (2003). “Religion and Politics:
Contention and Consensus — Growing Number Says Islam Encourages Violence
Among Followers.” Retrieved ﬁom: httpzl/people-
press.org/reports/display.php3?ReportID=189 (March 20, 2007).

25 Corwin E. Smidt, “Religion and American Attitudes Toward Islam and an
Invasion of Iraq,” Sociology of Religion, 66 (fall 2005): 243-262.

26 Voice of America. “Negative View of Islam Growing in U.S.” Retrieved ﬁ'om:
http://www.voanews.com/english/archive/2006-03/Negative-View—of-Islam-
Growing-in-US.cfm?CFID=56780432&CFTOKEN=97394999 (March 30, 2007).

27 Encyclopedia Britannica, Book of the Year 2004 (London: The Encyclopedia
Britannica Company, Ltd., 2004).

28 U.S. Census 2000 Special Reports: “We the People of Arab Ancestry.”
29 The Arab-American Institute. “Demographics.”

3° Guido H. Stempel III, "Where People Really Get Most of Their News,"
Newspaper Research Journal, 12 (fall 1991): 2-9.

3‘ Shoemaker and Reese, Mediating the Message.

32 Emmet Kennedy, “‘Ideology’ ﬁorn Destutt De Tracy to Marx,” Journal of the
History of Ideas, 40 (3) (July 1979): 353-68; David M. Hart (2002). Destutt de
Tracy: Annotated Bibliography. Retrieved ﬁ'om:
http://www.econlib.0rg/libraryfIracy/DestuttdeTracyBio.html (Nov. 21, 2008);
John D. Cash, Identity, Ideology and Conflict: The Structuration of Politics in
Northern Ireland. (Cambridge: Cambridge University Press, 1996); Jonathan
Dancy and Ernest Sosa, A Companion to Epistemology. (Oxford: Blackwell
Publishing, 1992); Michael Billig, Ideology and Opinions: Studies in Rhetorical

103

 

Psychology. (London : Sage Publications, 1991); Eliezer B. Rafael, Sociology and
Ideology. (Leiden and Boston: Brill Academic Publishers, 2003).

33 Terry Eagleton, Ideology: An Introduction. (Nee York: Verso, 1994).

34 Raymond Williams, Marxism and Literature. (New York: Oxford University
Press, 1977).

35 Samuel Becker, “Marxist Approaches to Media Studies: The British
Experience,” Critical Studies in Mass Communication, 1 (1984): 66-80.

36 Raymond Williams, Marxism and Literature. (New York: Oxford University
Press, 1977).

37 Alvin Gouldner, The Dialectic of Ideology and Technology: The Origin of
Grammar, and Future of Ideology. (New York: Seabury Press, 1976).

38 Shoemaker and Reese, Mediating the Message.
39 Shoemaker and Reese, Mediating the Message.
40 Shoemaker and Reese, Mediating the Message.
41 Shoemaker and Reese, Mediating the Message.
42 Shoemaker and Reese, Mediating the Message.
43 Shoemaker and Reese, Mediating the Message.

44 Stuart Hall, “Ideology,” in E. Barnouw (ed.), International Encyclopedia of
Communication Vol. 2. (New York: Oxford University Press, 1989).

45 Shoemaker and Reese, Mediating the Message.
46 Shoemaker and Reese, Mediating the Message.

47 Earl Babbie, The Practice of Social Research, 10th Edition. (Belmont, CA:
Wadsworth, 2004).

48 Babbie, The Practice of Social Research.
49 Babbie, The Practice of Social Research.

50 Babble, The Practice of Social Research.

104

 

5' Shoemaker and Reese, Mediating the Message.

52 Shoemaker and Reese, Mediating the Message.

53 Shoemaker and Reese, Mediating the Message.

54 Shoemaker, “Media Treatment of Deviant Political Groups.”

55 Shoemaker, Chang, Brendlinger, “Deviance as a Predictor of Newsworthiness.”
56 Shoemaker, Chang, Brendlinger, “Deviance as a Predictor of Newsworthiness.”
57 Shoemaker and Reese, Mediating the Message.

58 Edward L. Wells, “Theories of Deviance and the Self-Concept,” Social
Psychology, 41 (1978): 197-197.

59 Shoemaker and Reese, Mediating the Message.

60 Shoemaker and Reese, Mediating the Message.

6‘ Hallin, The Uncensored War.

62 Hallin, The Uncensored War.

63 Hallin, The Uncensored War.

64 Norman R. Yetman, Majority and Minority: The Dynamics of Race and
Ethnicity in American Life, 6th Edition. (Needham Heights, MA: Allyn & Bacon,
1999)

65 Yetrnan, Majority and Minority.

66 Phillip Gleason, “Minorities (Almost) All: The Minority Concept in American
Social Thought,” American Quarterly, 43 (3) (September, 1991).

67 Gunnar Myrdal, An America Dilemma: The Negro Problem and Modern
Democracy. (New York: Harper & Row, 1944).

68 Yetman, Majority and Minority; Gleason, “Minorities (Almost) All.”
69 Yetrnan, Majority and Minority.

7° Yetrnan, Majority and Minority; Charles Wagley and Marvin Harris, Minorities
in the New World. (New York: Columbia University Press, 1958); Robin M.

105

 

Williams, Strangers Next Door: Ethnic Relations in American Communities.
(Englewood Cliffs, NJ: Prentice Hall, 1964).

7' John E. Farley, Majority-Minority Relations, SmEdition (Upper Saddle Rive,
NJ: Pearson Education Inc., 2005).

72 Keever, Martindale and Weston, U.S. News Coverage of Racial Minorities.

73 Serge Moscovici, “Toward a Theory of Conversion Behavior,” in L. Berkowitz
(ed.), Advances in Experimental Social Psychology, vol. 13 (New York:
Academic Press, 1980).

74 Shoemaker and Reese, Mediating the Message.
75 Shoemaker and Reese, Mediating the Message.

76 Warren Breed, “Social Control in the Newsroom: A Functional Analysis,”
Social Forces, 33 (1), (May 1955).

77 Breed, “Social Control in the Newsroom.”

78 Shoemaker, Change and Brendlinger, “Deviance as a Predictor of
Newsworthiness”; Shoemaker, “The Perceived Legitimacy of Deviant Political
Groups”; McLeod, “Communicating Deviance”; Shoemaker, “Media Treatment
of Deviant Political Groups”; Shoemaker, Danielian and Brendlinger, “Deviant
Acts, Risky Business, and U.S. Interests”;

Shoemaker, “The communication of deviance”; Craig Murray, Katy Parry, Piers
Robinson, and Peter Goddard, “Reporting Dissent in Wartime: British Press, the
Anti-War Movement and the 2003 Iraq War,” European Journal of
Communication, 23 (March 2008): 7- 27; Jong Hyuk Lee, “Effects of News
Deviance and Personal Involvement on Audience Story Selection: A Web-
Tracking Analysis,” Journalism and Mass Communication Quarterly, 85 (spring
2008): 41-61; David Pritchard and Karen D. Hughes, “Patterns of Deviance in
Crime News,” Journal of Communication, 47 (3) (summer 1997): 49-67; Gabriel
Weirnann and Hans-Bernd Brosius, “The Newsworthiness of International
Terrorism,” Communication Research, 18 (3) (June 1991): 333- 54.

79 Shoemaker, “The Perceived Legitimacy of Deviant Political Groups.”
80 McLeod, “Communicating Deviance.”
8' McLeod, “Communicating Deviance.”

82 Lee, “Effects of News Deviance.”

106

 

83 Shoemaker, “Media Treatment of Deviant Political Groups.”

84 Weirnann and Brosius, “The Newsworthiness of International Terrorism.”

85 Pritchard and Hughes, “Patterns of Deviance in Crime News.”

86 Pritchard and Hughes, “Patterns of Deviance in Crime News.”

87 Pritchard and Hughes, “Patterns of Deviance in Crime News.”

88 Pritchard and Hughes, “Patterns of Deviance in Crime News.”

89 Michael J. Breen, “A Cook, a Cardinal, his Priests, and the Press: Deviance as a
Trigger for Interrnedia Agenda Setting,” Journalism and Mass Communication
Quarterly, 74 (2) (summer 1997): 348-56.

90 Breen, “A Cook, a Cardinal, his Priests, and the Press.”

91 Murray, Parry, Robinson, and Goddard, “Reporting Dissent in Wartime.”

92 Murray, Parry, Robinson, and Goddard, “Reporting Dissent in Wartime.”

93 Murray, Parry, Robinson, and Goddard, “Reporting Dissent in Wartime.”

94 Murray, Parry, Robinson, and Goddard, “Reporting Dissent in Wartime.”

95 Murray, Parry, Robinson, and Goddard, “Reporting Dissent in Wartime.”

96 Shoemaker and Reese, Mediating the Message.

97 Edward Said, Orientalism. (New York: Vintage, 1979).

98 Said, Orientalism.

99 Edmund Ghareeb (ed), Split Vision: The Portrayal of Arabs in the American
Media. (Washington DC: American-Arab Affairs Council, 1983); Michael W.
Suleiman, The Arabs in the Mind of America. (Brattleboro, VT: Amana Books,
198 8); Janice J. Terry, Mistaken Identity: Arab Stereotypes in Popular Writing.
(Washington DC: American-Arab Affairs Council, 1985).

99 Dina Gavrilos, “Arab Americans in a Nation’s Imagined Community: How
News Constructed Arab American Reactions to the Gulf War,” Journal of
Communication Inquiry, 26 (autumn 2002): 426-453.

'00 Gavrilos, “Arab Americans.”

107

 

'01 Gavrilos, “Arab Americans.”

102 David Domke, Philip Garland, Andre Billeaudeaux, John Hutchenson,
“Insights into U.S. Racial Hierarchy: Racial Proﬁling, News Sources, and
September 11,” Journal of Communication, 53 (winter 2003): 606-623.

'03 Domke, Garland, Billeaudeaux, Hutchenson, “Insights Into U.S. Racial
Hierarchy.”

'04 Mary A. Weston, “Post 9/11 Arab-American Coverage Avoids Stereotypes,”
Newspaper Research Journal, 24 (winter 2003): 92-106.

105 Steven Salaita, “Ethnic Identity and Imperative Patriotism: Arab Americans
EridOACﬁer 9/11,” College Literature, 32 (spring 2005): 146-169.

‘06 Salaita, “Ethnic Identity.”

‘07 Salaita, “Ethnic Identity.”

'08 Schur, Coming to America.

'09 Eric Hooglund (ed.), Taking Root: Arab-American Studies. (Washington DC:
The American-Arab Anti-Discrimination Committee, 1985.

”0 Schur, Coming to America.

111 James Zogby (ed.), Taking Root: Bearing Fruit. (Washington DC: The
American-Arab Anti-Discrimination Committee, 1984).

”2 Zogby, Taking Root.
”3 Zogby, Taking Root.
”4 Zogby, Taking Root.
”5 Schur, Coming to America.
”6 Schur, Coming to America.
”7 Zogby, Taking Root.

”8 The Arab-American Institute. “Demographics.”

108

 

”9 The Arab-American Institute. “Demographics.”

'20 The Arab-American Institute. “Demographics.”

121 The Arab-American Institute. “Demographics.”

'22 The Arab-American Institute. “Demographics.”

‘23 The Arab-American Institute. “Demographics.”

'24 Andrej Kulczycki, Arun Peter Lobo, “Deepening the Melting Pot: Arab-
Americans at

the Turn of the Century,” The Middle East Journal, 3 (summer 2001): 459-473.
125 The Arab-American Institute. “Demographics.”

'26 The Arab-American Institute. “Demographics.”

‘27 The Arab-American Institute. “Demographics.”

'28 The Arab-American Institute. “Demographics.”

129 The Arab-American Institute. “Demographics.”

‘30 The Arab-American Institute. “Demographics.”

13 l The Arab-American Institute. “Demographics.”

‘32 The Arab-American Institute. “Demographics.”

‘33 U. S. Census: “Combined Statistical Area Population Estimates File for Internet
Display,” 2006. Retrieved from: http:

www. census. gov/population/estimates/metro_general/2006/CSA- EST2006-
alldata. csv (January, 20, 2009)

‘34 U.S. Census: “Combined Statistical Area Population Estimates.”

'35 The Arab-American Institute. “Demographics”; U.S. Census: “Combined
Statistical Area Population Estimates.”

'36 Nabeel Abraham and Andrew Shryock (ed.), Arab Detroit: From Margin to
Mainstream. (Detroit: Wayne State University Press, 2000).

'37 U.S. Census: “The Arab Population: 2000,” December 2003. Retrieved from:
http://www.census.gov/prod/2003pubs/c2kbr-23.pdf (March 30, 2007).

109

 

‘33 Abraham and Shryock, Arab Detroit.

‘39 The Arab-American Institute. “Demographics.”

'40 The Arab-American Institute. “Demographics.”

'41 The Arab-American Institute. “Demographics.”

'42 The Arab-American Institute. “Demographics.”

'43 Abraham and Shryock, Arab Detroit.

”4 City Demographics: “Detroit Metropolitan Statistical Area,” 2002- Retrieved
ﬁ‘om: http://www.cityrating.com/citystats.asp?city=Detroit&state=MI (January
20, 2009).

‘45 City Demographics: “Detroit Metropolitan Statistical Area.”

'46 The Arab-American Institute. “Demographics”; City Demographics: “Detroit
Metropolitan Statistical Area.”

147 Abraham and Shryock, Arab Detroit.
148 Abraham and Shryock, Arab Detroit.
”9 Abraham and Shryock, Arab Detroit.
150 Abraham and Shryock, Arab Detroit.
‘5 ' Shoemaker and Reese, Mediating the Message.
'52 Shoemaker and Reese, Mediating the Message.
‘53 Shoemaker and Reese, Mediating the Message-
154 Shoemaker and Reese, Mediating the Message.
155 Shoemaker and Reese, Mediating the Message.
'56 The Arab-American Institute. “Demographics.”

‘57 The Arab-American Institute. “Demographics.”

110

 

'58 The Arab-American Institute. “Demographics”; City Demographics: “Detroit
Metropolitan Statistical Area”; U.S. Census: “Combined Statistical Area
Population Estimates.”

159 The Arab-American Institute. “Demographics”; City Demographics: “Detroit
Metropolitan Statistical Area”; U.S. Census: “Combined Statistical Area
Population Estimates.”

‘60 The Arab-American Institute. “Demographics.”
16‘ Shoemaker and Reese, Mediating the Message.
'62 Editor & Publisher International Yearbook
163 Editor & Publisher International Yearbook

'64 Daniel Riffe, Stephen Lacy and Frederick G. Fico, Analyzing Media Messages;
Using Quantitative Content Analysis in Research. (Mahwah, NJ: Lawrence
Erlbaurn Associates, 2005).

165 Cleveland G. Wilhoit and David Weaver, “Foreign News Coverage in Two
U.S. Wire Services: An Update,” Journal of Communication, 33 (2), (spring
1983): 132-149; Bob Franklin (eds), Local Journalism and Local Media.
(London: Routledge, 2006); K. M. Shrivastava, News Agencies, ﬁom Pigeon to
Internet. (New Delhi: New Dawn Press, 2007).

‘66 Rodney Stark and William Sims Bainbridge, Religion, Deviance, and Social
Control.(New York: Routledge, 1997); Helen Rose Fuchs Ebaugh, Handbook of
Religion and Social Institutions. (New York: Springer-Verlag, 2005); Mike
Hepworth, Confession: Studies in Deviance and Religion. (Boston: Routledge &
Kegan Paul, 1982); Robin D. Pertin, “When Religion Becomes Deviance,”
Teaching Sociology, 29, (April 2001): 134-152.

167 Paul C. Higgins and Richard R. Butler, Understanding Deviance. (New York:
McGraw-Hill, 1982); Allen E. Liska, Perspectives on Deviance. (Englewood:
Prentice-Hall, 1987).

168 Walter Lippman, Public Opinion. (New York: The Macmillan Company,
1922); Michael W. Suleiman, “American and the Arabs: Negative Images and the
Feasibility of Dialogue,” in Arab Americans: Continuity and Change, eds. Baha
Abu-Laban and Michael W. Suleiman (Belmont, Mass: Association of Arab-
American University Graduates, Inc., 1989); Said, Orientalism; Weston, “Post
9/11 Arab-American Coverage.”

111

 

'69 Guido H. Stempel III and Bruce H. Westley (eds.), Research Methods in Mass
Communication. 2"d Ed. (Englewood Cliffs, NJ: Prentice Hall, 1989).

170 Thomas W. MacFarland. “Spearman’s Rank-Difference Coefﬁcient of

Correlation.” Retrieved from:
http://www.nyx.net/~tmacfarl/STAT_TUT/spearman.ssi (July 28, 2009).

‘71 Stempel and Westley, Research Methods in Communication.
'72 Stempel and Westley, Research Methods in Communication.
'73 Stempel and Westley, Research Methods in Communication.
‘74 Shoemaker and Reese, Mediating the Message.
‘75 Shoemaker and Reese, Mediating the Message.
”6 Shoemaker and Reese, Mediating the Message-
”7 Shoemaker and Reese, Mediating the Message.
'78 Shoemaker and Reese, Mediating the Message.
'79 Shoemaker and Reese, Mediating the Message.
‘80 Shoemaker and Reese, Mediating the Message.
‘81 Shoemaker and Reese, Mediating the Message.
‘82 Shoemaker and Reese, Mediating the Message.

183 Society of Professional Journalists. “Code of Ethics.” Retrieved from:
http://www.spj.org/ethicscode.asp (July 27, 2009). i

184 The Associated Press. “The Associated Press Statements of News Values and
Principles.” Retrieved from: http://www.ap.org/newsvalues/index.html (July 27,
2009)

'85 Breed, “Social Control in the Newsroom.”
‘86 Pew Project for Excellence in Journalism. “The State of the News Media 2009;
An annual Report on American Journalism.” Retrieved from:

http://www.stateofthemedia.org/2009/narrative_newspapers_intro.php?media=4
(July 27, 2009).

112

 

'87 Pew Project for Excellence in Journalism. “The State of the News Media.”

'88 Pew Project for Excellence in Journalism. “The State of the News Media.”

113

 

”11111311111111ﬁll/ﬂilﬂllljlﬂllll“