r. mm»...

.3“. .‘ vyﬁi
.. . 25.8.5
3.3., .1;

.3 1 q
:2».- v a
ﬂu... :5

‘v

a‘fw =3 3

, ah. ‘. ”’ :1"
‘IQ‘OZ‘ENI 33-2133» ‘

15...:
‘1‘ «l

3.7

 

 

M

 

1.2-4an

mum“

mmm

{t

n .\ .
I.

44.... . ...-....,.n ,
$5»; am_n.§....a ., ,_ , . . u in? .1, . r ,
. . H: are. .1. .s $.63. -_...mm.hwm.§é§

 

This is to certify that the
dissertation entitled

NEWS MEDIA FRAMING OF THE IRAQ ISSUE IN THE
2004 US. PRESIDENTIAL CAMPAIGN

presented by

 

ARVIND DIDDI

has been accepted towards fulﬁllment
of the requirements for the

Ph.D. degree in Mass Media

Major Professor’s Sig‘n’ature
A’du, l3 Lgcaé

Date

 

 

... o -o-l-I-0-0-O-I-0-0-O-4-‘-.-'-I-0-0-U-l-O-0-0-0-0-0-I-I-.-‘-'-'

 

MSU is an Afﬁnnative Action/Equal Opportunity Institution

ﬁfe. w

 

 

LIBRARY
Michigan State
University

 

 

 

 

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

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2/05 p:lClRC/DateDue.indd-p.1

 

 

NEWS MEDIA FRAMING OF THE IRAQ ISSUE IN THE 2004 US.
PRESIDENTIAL CAMPAIGN

By

Arvind Diddi

A DISSERTATION
Submitted to
Michigan State University
In partial fulﬁllment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY
Mass Media Ph.D.

School of Journalism

2006

ABSTRACT

NEWS MEDIA FRAMING OF THE IRAQ ISSUE IN THE 2004 US.
PRESIDENTIAL CAMPAIGN

By
Arvind Diddi

This study content analyzed 445 stories from three networks and three
cable channels to examine the news media framing of the Iraq issue in the 2004
US. presidential campaign. In order to examine how various factors inﬂuence the
structural qualities of news in terms of framing, this study assessed the inﬂuence
of nature of medium (networks vs. cable), nature of journalists (race and gender)
and nature of sources (partisans vs. non-partisans) on the Iraq issue frames.
Derived from the past literature, twelve frames were deﬁned apriori for this study.

The data of the study revealed that the news media largely emphasized
negative frames sponsored by Bush and Kerry partisans. Overall, stories with
dominant frames were more likely to favor Kerry than Bush. Journalists usually
linked candidate partisan frames with a political fight frame, which emphasized
the “horse race” and how candidates were using the Iraq issue for electoral gain.

The hierarchical model used in this study was robust in predicting the
inﬂuence of partisan sources on news media framing of the Iraq issue. It was
moderately successful in showing framing differences based on medium and
journalist gender (but not race) on the Iraq issue frames. It replicated past work
showing the inﬂuence of partisans not only broadly on the news, but speciﬁcally

on public policy issue framing.

To resolve the theoretical debate on generic versus issue-speciﬁc frames,
this study suggests that issue-speciﬁc frames should be anchored in generic
frames such as conﬂict, responsibility, and human interest, etc. Anchoring issue
speciﬁc frames into generic frames would facilitate cross-issue generalizability,
thereby making framing studies more robust in their prediction, explanation, and

control of a phenomenon in mass communication.

Dedicated to My Parents

iv

ACKNOWLEDGEMENTS

In the Indian culture and tradition, guru—the teacher— is equated with the
holy trinity. I feel blessed and fortunate for being able to ﬁnd such divine force in
Dr. Frederick F ico, chairperson of my dissertation committee, and Dr. Steve Lacy,
Dr. Lucinda Davenport, and Dr. Charles Sahnon, my dissertation committee
members. Thank you, Dr. F ico, for your constant support, guidance, and
encouragement through out the project. Your expertise helped me to turn a ﬂood
of information into a manageable effort. Dr. Lacy, Dr. Davenport, and Dr.
Salmon, your advice and direction were immensely helpful. Without your
individual and collective support and guidance this project would have been much
more difﬁcult to manage. For your time and efforts on my behalf, I thank you.

Additionally, this project would not have been possible without the love,
support, and encouragement of my family. I thank my mother, father, sister and
brother for all they have done to enable me to continue my journey. I also thank
all faculty, staff and colleagues at journalism school, and my friends at apartment

C-48 in East Lansing for their constant support and affection.

TABLE OF CONTENTS

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

LIST OF TABLES ix
LIST OF FIGURES x
CHAPTER 1: Framing and The Gulf War 1
Introduction 1

Focus of the Study 2

The Iraq War and the 2004 Presidential Race 4
CHAPTER 2:Types of Frames and Their Sources 9
Theoretical Framework _ 9
Literature Review on Issue Framing 12

Iraq Issue Framing 17

Gulf War Report Studies: The Military Emphasis 18

Studies on the Policy Conﬂict Over the Iraq War ................. 20

Relevant Iraq War Frames in the 2004 Presidential Race 22
Factors Inﬂuencing News Media Iraq Issue Frames 28
Nature of the Medium ...................................................... 31

Race and Gender ofJournalist...........................................35

Partisan Balance of Storres40

CHAPTER 3: Linkages of Frames to Sources 44
Research Questions and Hypotheses 42

A Model of Inﬂuences on Framing 49
CHAPTER 4: Content Analyzing Iraq War Frames ‘1
Method ‘1
Sample - <1
Coding Procedure 52
Measures ..................................................................... 53

Dependent Variable: Iraq Issue Frames .............................. 53

Reliability and Validity ‘8
Independent Variable Measures . ....................................... 60

 

vi

 

 

 

 

 

 

 

 

 

Data Analysis 62
Rules for Assessing Research Questions and Hypotheses .......... 63

Rule Application to Research Questions and Hypotheses...........64
Assessing Relative Strength of Inﬂuences on Framing ............. 66
CHAPTER 5: Who Framed the War and How 68
Results 68
Description of Data by the Number of Stories 68
Description of Data by the Time Devoted to Stories and Frames .......... 72
Research Questions and Hypotheses 75
Framing Patterns in Coverage ......................................... 75
Network and Cable TV Difference in Framing 86
Journalists and Their Frames .......................................... 89
Partisans and Framing .................................................. 98
Non-partisans and Framing....................... 102
Summary of Research Questions and Hypothesis Test Results.....104
Assessment of Relative Strength of Inﬂuences ............................. 106
Chapter 6: Discussion 110
Implications for Theory 112
Reassessing The Model of This Study .............................. 112

Using Generic and Issue Specific Frames ......................... 114
Revised Model of Inﬂuences on Framing .................................. 119
Predictors of Media Framing - - 120
Framing Power of Journalists and Sources 124

 

vii

Links Between Framing and Other Theoretical Approaches ................ 127
Implications for Journalism .................................................. 130
Resource Commitment ............................................... 130
Journalists’ Socialization and T raining...............................131
Fairness and Balance132
Quality Control and Accountability 133
Implications for Research ...................................................... 135
Applying Generic/Issue Specific F rames.... 135

Varying Media Type and Frames 135

Varying News Gatherer Resources and Framing ......... 136
Varying News Gatherer Diversity and Framing ........... 136
Male/Female Ratio and Framing ............................. 137

Content Analysis and its Limitations in Framing Researchl37

Socially Responsible Reporting and Scholarly Research..l38

 

 

Appendix 1: Content Analysis Protocol ...... 140
Appendix II: Frame Type Examples 151
References- - - - 159

 

viii

Table 1

Table 2

Table 3

Table 4

Table 5

Table 6

Table 7

Table 8

Table 9

Table 10

Table 11

Table 12

Table 13

Table 14

Table 15

Table 16

Table 17

Table 18

Table 19

Table 20

Table 21

LIST OF TABLES
Percentage of the Iraq Issue Stories in Television Channels....68
Percentages of Types of Stories ..................................... 69

Percentages of Story Types by Networks and Cable Channels70

 

Race and Gender of Journalists and Story Percentages ......... 71
Number of Stories With Non-partisan Sources ................... 72
Percentages of Story and Framing Seconds 72
Type of Story and Percentage of Time 73

 

Race and Gender of Journalists and Framing Seconds

 

 

 

 

Percentages 74
Non-partisan Framing Seconds and Stories 74
Dominant Frames in Stories 76
Frames in News Stories 77
Percentages of Total Story Seconds and Framing Seconds ....... 79

Percentages of Iraq Issue Framing Seconds in News Channels 84

Percentages of Framing Seconds in Network and Cable

 

 

 

 

Channels 87
Frequently Used Frames by Journalists _- 91
J oumalists’ Race and Framing Seconds 95
Journalists’ Gender and Framing Seconds 98
Partisan Story Domination and Percentage of Frames ............. 100

Percentages of Republican and Democratic Sourced Frames ...101

Frames Used by Expert/Non-partisan Sources ""103

 

Inﬂuence of Factors on Partisan Frame Imbalance ................... 109

ix

LIST OF FIGURES
Figure 1 A Model of Inﬂuences on Framing 49

Figure 2 Revised Model of Inﬂuences on Framing 119

 

CHAPTER 1: Framing and The Gulf War
Introduction

This study examines the framing of the Iraq issue in television news
during the 2004 US. presidential campaign. Framing literature in mass
communication and political communication emphasizes that it is not only the
issues of the day (McCombs, Shaw, & Weaver, 1997; Wanta, 1997), but how
those issues have been covered (i.e., framed) in mass media that play a crucial
role in inﬂuencing public opinion (e.g., Dirnitrova, Kaid, Williams, & Trammell,
2005; Entrnan, 1991, 1993; Gitlin, 1980; Iyengar, 1991; Luther & Miller, 2005;
Rhee, 1997; Shoemaker & Reese, 1996). Some scholars (e.g., McCombs, Llamas,
Lopez-Escobar, & Rey, 1997; Ghanem, 1997) argue that framing is an extension
of agenda-setting theory and refer to framing as second-level agenda setting. They
suggest that attribute salience or media frames of issues or events are transferred
from the media agenda to the public agenda (but see Scheufele [2000] for a
contrary view).

The way an issue is covered by the mass media may sway public opinion
in favor or against that issue and important people associated with it. Framing of
issues by the mass media in general and news media in particular attains a greater
signiﬁcance during presidential campaigns as rival political parties and candidates
use them to sponsor their issue positions (frames) in order to garner public
support (e. g., Allen, O’Loughlin, Jasperson, & Sullivan, 1994).

Framing is inherent to journalism because journalists gather, organize, and

present news in a simple and comprehensible form to their mass audiences

(D’Angelo, 2002; Scheufele, 1999; Valkenburg, Semetko, & de Vreese, 1999).
The way a news story is presented, or in other words framed, by a journalist
depends on several factors such as sources, news media routines, the nature of the
outlet, organizational pressures, and journalists’ ideology (Shoemaker & Reese,
1996). However, this does not mean that news media are the sole players in
inﬂuencing public opinion on some issue. Instead they are the key players among
other channels such as political ads, and public debates (Holbert, Benoit, Hansen,
& Wen, 2002).

Focus of the Studv

This study examines framing of the Iraq issue in the 2004 US. presidential
campaign between Labor Day (September 6, 2004) and the day of election
(November 2, 2004) for several reasons. Theoretically, this study examines the
process of the formation of news media frames. In other words, it assesses the
inﬂuence of communication variables such as the nature of the medium, sources,
and journalists, on the formation of news media frames. As Scheufele (1999)
argues, previous research has examined the process of creation of news (Gans,
1980; Tuchrnan, 1978), and similar studies in framing are needed to understand
the process of creation of news media frames, which are often a product of a
complex interaction between the journalists and factors such as nature of the
medium, sources, news media routines, and organizational pressures (Shoemaker
& Reese, 1996).

Framing analysis allows a deeper and broader understanding of an issue

and also gives insight into the power play in the prevailing socio-political

environment. This news media framing study examines the practice of journalism
in America, which according to some scholars (Aday, Livingston, & Hebert,
2005; Entman, 2004; Hutcheson, Domke, Billeaudeaux, & Garland, 2004; Norris,
Kern, & Just, 2003; Ross & Bantirnaroudis, 2006; Scheufele, Nisbet, & Ostman,
2005; Williams, 2003; Zelizer & Allan, 2002) has undergone noticeable change
after the September 11, 2001 attacks. For instance, Entman (2004) argues that
through 19805 American journalists used to operate under a Cold War paradigm,
which disappeared aﬁer the disintegration of the Soviet Union. According to
Entrnan (2004), the September 11 attacks provide a broader context to understand
the role of the media. Thus, this study examines the Iraq issue in the context of a
presidential election campaign, in which other institutions deliberately seek to
inﬂuence media coverage for policy and political ends. This study, therefore, has
a different focus from others that have analyzed the Iraq issue in the context of the
war itself (e.g., Aday, Livingston, & Hebert, 2005; Allen et al., 1994; Dimitrova
et al., 2005; Fico, Ku, & Sofﬁn, 1994; Hiebert, 2003; Luther & Miller, 2005;
Reese & Buckalew, 1995).

Substantively, the Iraq issue was chosen because it emerged as one of the
most important issues during the presidential campaign (Patterson, 2004; Pew,
2004), and has major socio-political implications for America. The Iraq issue has
several potential dimensions to its framing, including the threat of weapons of
mass destruction, war on terrorism, establishing democracy in Iraq, civil liberties
in America and prisoner-abuse abroad, the implications of withdrawal of

American troops, American foreign policy, and economic costs of the war. Also,

there is a fundamental difference between the ﬁrst and second Gulf wars. Among
other political reasons, the 1991 Persian Gulf War was prompted by the initiative
to liberate Kuwait from Iraqi occupation. The 2003 Iraq war was primarily fought
in the background of the September 11 attacks, but soon acquired an even
broader, historical context. Literature suggests that after the 19603, whenever
America got involved in a conﬂict requiring deployment of its troops abroad for a
longer duration, it raised the question of ‘whether this will be another Vietnam?’
(e.g., Entman, 2004; Gartner, Segura, & Wilkening, 1997; Muller, 1973).

While the Iraq issue is multidimensional and has policy implications
beyond elections, this study focuses on the way this issue was covered by the
American television news—ABC, NBC, CBS, CNN, FOX, and MSNBC—during
the presidential campaign. The Iraq issue was crucial for voters’ decisions on
whether to continue with the presidency of George W. Bush or elect his
Democratic Party opponent, John Kerry. The Iraq issue gave ample opportunity to
both the contenders to take speciﬁc positions and promote their stand through
news media and other avenues such as political ads, political debates, and public
meetings. Therefore, this study assesses the news media’s role in covering the
Iraq issue in the presidential campaign and examines the inﬂuence of various
communication variables on the creation of news media frames. A free and fair
news media is essential for a democracy to ﬁrnction, and mass media scholars _
perform their own watchdog function on news media performance through

critical, theoretical and empirical studies of media coverage.

The Iraq War and the 2004 Presidentialﬁrgg

Before delving into theoretical aspects of this study, a brief background on
the Iraq war and the run-up to presidential race can provide context for the
importance of such scholarly attention. The day after the September 11, 2001
terrorist attacks on American soil, President George W. Bush addressed the nation
and called for people to unite against the war on terror (Entman, 2004). In about a
month, US. troops led a coalition to dismantle the Taliban regime in Afghanistan
for sheltering Osama bin Laden, the leader of Al Qaeda, and mastermind of
September 11 attacks.

The next step in the war against terror saw Iraq identiﬁed as one of the
“axis of evil” nations. The Bush administration argued that there was a connection
between Saddam Hussein, Osama bin Laden, and the September 11 attacks. The
Bush administration also argued that there was an immediate need to remove
Saddam Hussein from power because he possessed weapons of mass destruction
and constituted a potential threat to American security (Danchev & MacMillan,
2005). President Bush sought Congressional approval to invade Iraq and was
successful in getting the resolution passed in October 2002. The Senate gave
approval by 77 to 23 and the House by 296 to 133. Republicans overwhelmingly
supported Bush (48-1 in the Senate and 215-6 in the House), while Democrats
were more divided in their support (29-22 in the Senate), or even opposed (81-127
in the House) (Congressional Quarterly, 2003). This congressional approval for

invading Iraq was greater in the second Gulf War when compared to the ﬁrst

Persian Gulf War in 1991, when the Senate voted 52 to 47 in favor, and the House
250 to 183 (Aday, Cluverius, & Livingston, 2005).

President Bush on March 17, 2003 gave an ultimatum to Saddam Hussein
to leave Iraq within 48 hours. After the March 19 announcement by Bush that
troops were engaged in Iraq, the Senate (unanimously) and the House (392-11)
passed resolutions backing the commander-in-chief (Dumbrell, 2005). The
“coalition of willing” of about forty nations led by the US. troops launched
“Operation Iraqi Freedom” on March 19, 2003. The war lasted for more than a
month. The Iraqi military was defeated and Baghdad fell on April 9, 2003.
President Bush on May 1, 2003 declared the end of major combat operations, and
Coalition forces ﬁnally captured Saddam Hussein on December 13, 2003 (Aday,
Cluverius, & Livingston, 2005; Denchev & MacMillan, 2005; Dimitrova et al.,
2005).

However, all along there was some opposition to the war from
Democrats and a few Republicans, who wanted emphasis on diplomatic and
multilateral solutions rather than use of force. And congressional support eroded,
especially among Democrats.

With regard to American public opinion on using force against Iraq, a
Gallup poll released in middle of March 2003 indicated 88% of respondents
ﬁrmly believing that Iraq backed terrorist organizations with plans to attack the
United States. About 51% afﬁrmed that Saddam Hussein was connected to 2001
attacks (Schneider, 2003). In general, polls consistently showed public support for

Bush’s foreign policy. For instance, in a February 2003 Gallup poll 90% of

Americans saw Iraq as a threat to the United States, and in a CBS poll 66%
favored military action (Schneider, 2003a).

However, a closer examination of the public opinion pointed at certain
split in the support for war as people expressed concerns about going it alone in
Iraq, and dangers in the aftermath of war. Polling showed sharp partisan division
over the war in Iraq. For instance, in February 2003, a Gallup poll revealed that
29% of Democratic respondents opposed invasion of Iraq under any
circumstances. Another New York T imes/CBS poll carried out on the ﬁrst day of
the 2003 Iraq war revealed 93% of Republicans, but only 50% of Democrats,
supporting President Bush’s handling of the war. After President Bush announced
the official end to the Iraq war on May 1, a Newsweek poll in June 2003 showed
38% of respondents criticizing the administration for ‘purposely misleading’ the
public over weapons of mass destruction. Criticism of the war and its objectives
increased, again very much in line with partisan ties, as the US. casualties
increased (Dumbrell, 2005).

The mounting casualties of the US. troops and the increasing insurgency
in Iraq prompted Senator Kerry to term the Iraq war as a ‘wrong war, at a wrong
place, at a wrong time,’ a comment made earlier by Howard Dean during
Democratic primaries. Republicans reacted to the Kerry criticism by saying that
Senator Kerry lacked a consistent position on the Iraq issue and that he ﬂagrantly
pandered to the Democratic leﬁ when he borrowed lines from Howard Dean to

criticize the Iraq policy.

President Bush ultimately won reelection largely on the strength of his
national security and social issue positions. A post-election survey by Pew (2004)
revealed that 25% of respondents singled out Iraq as the most important issue of
the 2004 presidential race, while moral issues with 14% and economy/jobs with
12% took the place of second and third most important issues.

As the Iraq issue emerged as a complex and crucial issue inﬂuencing
public opinion, it is interesting to examine how the news media covered this issue
during an election, especially how it was framed in the media landscape by the
news outlets, journalists, and partisan and non-partisan sources. A study from the
mass media perspective is important because it has implications for public
opinion. Certainly the terrorist attack on the US, the subsequent U.S. invasions
and the rise of the Iraqi insurgency are objective facts that inﬂuence US. political
leaders and the broader public. But the US. media were a prism through which
these events were ﬁltered. And the US. presidential race further intensiﬁed how
those distant events inﬂuenced opinions and, ultimately, votes for President Bush
or Senator Kerry. The framing of the Iraq issue was the means by which those
distant events could ultimately be translated into opinions and then votes. Framing
helped provide the mental categories by which citizens could think about
problems, their causes, who was responsible, and what solutions could work to
solve those problems. The purpose of this study, then, is to illuminate part of that
conceptual process by identifying the framing of the Iraq issue frequently
advanced in television news, the sources of those frames and their public policy

content and implications.

CHAPTER 2: Types of Frames and Their Sources
Theoretical Framework

The empirical studies on framing can be broadly categorized into research
focusing on media frames and studies examining audience frames (Cappella &
Jamieson, 1997; D’Angelo, 2002; de Vreese, Peter, & Semetko, 2001; Entman,
1993; Scheufele, 1999). The emphasis of media frame research is on how issues
are covered and presented in the news (e.g., Andsager & Powers, 1999; Callaghan
& Schnell, 2001; de Vreese et al., 2001, Shah, Watts, Domke & Fan, 2002;
Norris, 1995; Semetko & Valkenburg, 2000). Audience frames research is
primarily concerned with how individuals interpret, perceive, and organize events
and issues (e.g., Brewer, 1992; de Vreese, 2004; Jacoby, 2000; Jasperson, Shah,
Watts, Faber & Fan, 1998; Nelson, Clawson, & Oxley, 1997; Price, Tewksbury,
& Powers, 1997; Rhee, 1997; Valkenburg, Semetko, & de Vreese, 1999). Some
research has examined both frames in the news and the effects of these frames on
audiences (e.g., Cappella & Jamieson, 1997; de Vreese, 2004; Iyengar, 1991;
Neuman, Just, & Crigler, 1992).

Citing the work of Cappella & Jarnieson (1997), de Vreese, Peter, and
Semetko (2001) emphasize that to increase “our understanding of effects of
frames, frames must have identiﬁable conceptual and linguistic characteristics and
be commonly observed in journalistic practice,” (p. 108). Therefore, content
analyses of news frames may be considered a crucial requirement prior to taking
up the study of effects of news frames. It does not make enough sense to

investigate the complex process of news consumption and framing effects with

frames that are “infrequent, insufﬁciently described or not a consistent component
of the news environment,” (Cappella & Jamieson, 1997, p. 49).

Framing research in news has often been carried out in relation to a
speciﬁc issue (e.g., Brewer, 2002; Callaghan & Schnell, 2001; Jasperson et al.,
1998; Norris, 1995) or a speciﬁc event (e.g., Entman, 1991; McLeod & Detenber,
1999; Mendelsohn, 1993; Rhee, 1997). A few studies have attempted to compare
the framing of certain issues or events across different media and news outlets
(Martin & Oshagan, 1997; Neuman et al., 1992; Rhee, 1997; Semetko &
Valkenburg, 2000; Simon & Xenos, 2000). A few studies have examined how
frames emerge and develop over time (e.g., Patterson, 1993), and some others
have compared framing of different issues (e.g., Neuman et al., 1992).

Framing cannot be limited by a single deﬁnition, but all of its deﬁnitions
underscore similar characteristics. According to Goffman (1974), the central
purpose of ﬁames is to help people “locate, perceive, identify, and label” the ﬂow
of information around them (p. 21). Tuchman (1978) argues that ﬁames help to
“narrow the available political alternatives” (p. 156). In their work, Neuman et al.
(1992) argue that news frames are “conceptual tools which media and individuals
rely on to convey, interpret and evaluate information,” ( p. 60). Tuchman (1978)
emphasized that they set parameters “in which citizens discuss public events” (p.
iv). Frames are “persistent selection, emphasis, and exclusion” (Gitlin, 1980, p.
7). Framing is selecting “some aspects of a perceived reality in such a way as to
promote a particular problem defmition, causal interpretation, moral evaluation,

and/or treatment recommendation,” (Entman, 1993, p. 53).

10

In response to Entman’s (1993) call for the establishment of a paradigm of
news framing research, D’Angelo (2002) argues that there should be no single
“mended” paradigm of framing research. The best way to beneﬁt framing
research is to use speciﬁc theories to progressively explicate a complex process.
By construing news framing as a research program, it is possible to clarify both
the practical conduct of scholarly inquiry and the real-life complexity of news
framing.

Continuing with the research line examining framing of issues, this study
attempts to analyze how the Iraq issue was framed in network and cable television
news between Labor Day (September 6, 2004) and election day (November 2,
2004). Speciﬁcally, in line with Scheufele’s (1999) argument, this study attempts
to examine media frames as a dependent variable to assess the inﬂuence of factors
extrinsic and intrinsic to media on the structural qualities of news in terms of
framing. Scheufele (1999) argues that: “Although many researchers have
examined extrinsic and intrinsic factors inﬂuencing the production and selection
of news (e.g., Gans, 1979; Shoemaker & Reese, 1996; Tuchman, 1978), no
evidence has yet been systematically collected about how various factors impact
the structural qualities of news in terms of framing,” (p. 109). He argued that
Tuchman’s (1978) work was more about the “‘act of making news’ (p. 12) rather
than framing of reality...Tuchman’s analysis revealed less about the relationship
between news organizations or journalistic norms and framing than could be a
more narrowly designed study,” (p. 110). While several factors may inﬂuence

news media frames, based on previous literature (e.g., Scheufele, 1999;

11

Shoemaker & Reese, 1996; Tuchman, 1978) this study attempts to examine the
inﬂuence of nature of the medium (network vs. cable), nature of the journalists
(race and gender), and nature of the sources (partisan, nonpartisan, and the
journalists themselves) on the framing of the Iraq issue in the 2004 presidential
race.

Literaere Review on Issue Frame_'_g

Studies on media frames basically deal with how issues are covered and
portrayed in the news (de Vreese et al. 2001; Entman, 1991; Semetko &
Valkenburg, 2000). In terms of the cognitive approach, frames exist as prior
knowledge, used by individuals to “efﬁciently” process information conveyed in
news frames (e.g., de Vreese, 2004; Shen, 2004; Rhee, 1997). Individuals also
discuss political issues using their mental frames about those issues (Gamson,
1992, 1996). In addition to the cognitive and interpersonal approach, ﬁames
operate within the routines and discourses of groups (Gitlin, 1980; Meyer, 1995).

Framing also shapes public opinion about political issues (D’Angelo,
2002). Framing researchers are also of the opinion that journalists self-
consciously play a “conduit” role (see Weaver& Wilhoit, 1986) in which they
make an attempt to “provide citizens with the information that will enable them to
gain an adequate understanding of politics,” (Schudson, 1983, p. 15). Thus,
framing researchers also monitor how well journalists perform this role (Hallin &
Mancini, 1984, p. 834).

To increase their power in a conﬂict, all political players, interest groups,

and politicians alike need the mass media to convey information to the general

12

public and to expand their visibility (Callaghan & Schnell, 2001; Cigler &
Loomis, 1995). Unlike direct communication between political players and the
citizens, information provided by the mass media is second-hand knowledge. For
instance, as Walter Lippmann (1920) explained it, the “pseudo-environment of
reports, rumors, and guesses,” (p.54). Therefore, the news media can have a
signiﬁcant inﬂuence over the ﬂow and shape of policy and issue debates. The
power to frame issues therefore may translate into the power to shape and win
policy debates.

de Vreese et a1. (2001) distinguished between issue-speciﬁc news frames
and generic news ﬁ'arnes. They argue issue-speciﬁc frames are applicable to
speciﬁc topics or news events, whereas generic frames are broadly applicable to a
range of different news topics, some even over time, and potentially, in different
cultural contexts. An issue-speciﬁc news frame allows for investigation of the
framing of particular events in greater detail. It may capture speciﬁc aspects of
selection, organization, and elaboration that are present in news coverage and
pertain speciﬁcally to a well-defmed issue. However, the greater detail and issue-
sensitivity renders issue speciﬁc analyses difﬁcult to generalize, capture, and use
as a base for general hypothesis and theory building (de Vreese et al., 2001). A
few examples of issue-speciﬁc framing studies include news framing of the ﬁrst
Gulf War (Reese & Buckalew, 1995), an investigation of the US. press and
television network coverage of two international airline accidents (Entman, 1991 ),
an analysis of the coverage of the US. national budget deﬁcit (Jasperson et al.,

1998), and presence of frames in Canadian television news coverage of the 1988

13

national Canadian election (Mendelsohn, 1993). Issue-speciﬁc news frame studies
are similar to an inductive approach involving analysis of a news story with an
open view to attempt to reveal the array of possible frames, beginning with
loosely deﬁned preconceptions of these frames (see for example, Gamson, 1992).

On the other hand, generic frames provide little scope for analyzing the
framing of an event or an issue in greater detail, but they facilitate comparisons
between frames, topics, and framing practices within and between different media
outlets (de Vreese et al., 2001). Several studies have recognized and underscored
the signiﬁcance of some frames in the news by focusing on their effect on
people’s interpretation of events and issues (Cappella & Jarnieson, 1997; Graber,
1988, 1993; Iyengar, 1991; Iyengar & Kinder, 1987; Neuman et al., 1992; Norris,
1995; Patterson, 1993). These studies and related literature have provided
important information about the occurrence or the effects of frames, and as of
now attempts have been made to develop a standard set of content analytic
indicators that can be used to reliably measure the prevalence of common ﬂames
in the news (see for example, de Vreese et al., 2001; Iyengar, 1991; Semetko &
Valkenburg, 2000; Shen, 2004; Valkenburg et al., 1999). “A reliable set of
content analytic indicators is necessary for studying developments in the news
over time and similarities and differences in the ways in which politics and other
topics of national and international importance are framed” in different news
outlets (Semetko & Valkenburg, 2000, p. 24).

In order to ﬁnd a common ground between issue-speciﬁc frames and

generic frames, and devising a framing approach with cross-issue generliazability,

14

Chyi and McCombs (2004) proposed a frame measuring scheme based on space
and time dimensions and referred to it as “frame-changing.” At the space
dimension, they suggested ﬁve levels—individual, community, regional, societal,
and intemational— which could be interpreted as intervals on a continuum. The
idea here is to examine an issue-frame at each of these ﬁve levels. The second
dimension, time, would enable to examine framing of an issue in the past and the
present, and on that basis predict how it could be framed in the future (Chyi &
McCombs, 2004).

Using this framing scheme across space and time dimensions, Chyi and
McCombs’ (2004) research ﬁndings revealed that an issue like Columbine school
shooting in 1999, at the space dimension, changed from being a individual-level
frame to societal frame. At the time dimension, the “present frame” emerged as a
dominant frame followed by “past” and “future” frames. In sum, they argued that
using this frame measuring scheme would allow cross-issue generalizability (Chyi
& McCombs, 2004).

All these framing approaches involve predefming certain frames as
content analytic variables to assess the extent to which these frames occur in the
news. In other words, “it is necessary to have a clear idea of the kinds of frames
likely to be in the news, because the frames that are not deﬁned apriori may risk
being overlooked. This approach also makes it necessary to measure frames in a
way that can be replicated, can cope with large samples, and can detect

differences in framing between media,” (e.g., electronic vs. print media) “and

15

within media,” (e.g. cable vs. network news) (Semetko & Valkenburg, 2000, p.
95).

Past research has identiﬁed a few frames that are commonly found in the
news, though not necessarily at the same time. Framing research, in general has
focused on the existence of one or more frames in the news and their implications
for public Opinion (Semetko & Valkenburg, 2000). The conﬂict ﬁ'ame, for
example, has been extensively debated (Cappella & Jamieson, 1997; Patterson,
1993), so also attribution of responsibility in the news (Iyengar, 1991). Other
studies by Neuman et al., (1992), Valkenburg et al., (1999), and Semetko and
Valkenburg (2000) went a step further as these researchers identiﬁed four
different frames that were common in US. news coverage of a range of issues,
including conﬂict, economic consequences, human impact, and morality frames.
Reese and Buckalew (1995), apart from the conﬂict frame, suggested the control
frame and the consensus frame in their study of coverage of Persian Gulf War in
local television. Their study suggested that conﬂict frame was used to emphasize
dissent between two unlike groups in the story, control ﬁame emphasized on
story’s reliance of ofﬁcial sources, and consensus ﬁame emphasized on
community solidarity. In a study of framing of the 2004 senate races, Love and
Fico (2006) examined newspaper coverage to identify the frames sponsored by
the reporters, partisan and nonpartisan sources. Citing previous literature (Kerbel,
Apee, & Ross, 2000; Rosen & Taylor, 1992; Trimble & Sampert, 2004), they
argued the importance of generic frames, especially the strategic frame in election

coverage. The strategic ﬁ'arne, tying into objectivity, emphasizes ‘horse race’

16

coverage, i.e. who is winning and losing, and reduces the chances of appearin
biased. However, a downside to the strategic frame is that it overlooks speciﬁc

issue positions and downplays how political players stake out clear issue positions

(Love & F ico, 2006)./
Iraq Issue Framing

Among several issues such as economy/jobs, health care, social security
and taxes, the Iraq issue along with moral values emerged as one of the top
concerns among voters in 2004 US. presidential race (Pew Research Center for
the People and the Press, 2004; Patterson 2004). Between Labor Day (September
6, 2004) and the day of election (November 2, 2004), the Iraq issue was one of the
prominent election topics covered in the news media. There was a considerable
political tension and also lack of consensus in public opinion on the way things
were going in Iraq. There were heated political debates on failure of intelligence
agencies in fmding weapons of mass destruction in Iraq. The Democratic party
and its supporters were highly critical of the Bush administration’s handling of
reconstruction of Iraq and conducting elections in that volatile region, which was
witnessing frequent conﬂicts between US. armed forces and local insurgents and
terrorists. Some voters with their relatives and ﬁiends serving in Iraq, anti-war
coalitions, the Democratic party, as well as those who strongly supported
President Bush, made Iraq a contentious issue during the election period. And
news media are known to cover contentious issues during ﬁnal stages of
campaign (Esser & D’Angelo, 2003). Thus, among other political issues, the Iraq

issue emerged as atop contender for news media coverage.

17

 

 

The assessment of Iraq War framing in the 2004 presidential election can
draw on a rich set of news content studies. Some of these studies focus on military
aspects of the ﬁrst Gulf War in 1991 and others on the 2003 war. Other studies
focus on the policy conﬂict that accompanied both wars. An examination of this
literature is crucial for determining the frames appropriately studied in this
research, and helps this research to appropriately replicate and extend past
research.

Gulf War Report Studies: The Military Emphasis

Several studies have analyzed the Persian Gulf War in 1991 in the context
of armed conﬂict and source dominance (e.g., Allen et al., 1994; Fico, Ku, &
Sofﬁn, 1994; Reese & Buckalew, 1995). Allen et al., (1994) in their study of the
ﬁrst Gulf War found that media not only served as conduits to military
information, but also framed and primed opinions of dissent, patriotism,
technology and elite consensus in way that subdued dissent and affected citizens’
evaluations of military actions. Reese and Buckalew (1995) examined local
television coverage of the Persian Gulf War and found that it linked the war
coverage to the media routines of television newswork, indicating how they act as
coherent frames supportive of Gulf policy. Their conﬂict frame projected anti-war
protest as being in opposition to patriotism. The controlframe highlighted protest
as a threat to social order. And the consensual frame linked Gulf Policy to
patriotism and ‘the troops,’(p.40). In a content study of newspaper coverage of the
United States in Gulf War, Fico, Ku, and Soffm (1994) examined how well the

press adheres to a socially responsible performance. Their study discovered anti-

18

war advocates being favored with more than six in ten stories overall dominated
by their assertions.

Studies on military and sourcing aspects of the second Gulf war focused
on the initial coverage of the conﬂict (e.g., Aday, Cluverius, & Livingston, 2005;
Aday, Livingston, & Hebert, 2005; Dimitrova et al., 2005). In their framing
analysis, Dimitrova et al., (2005) argued that the online news played a major role
during the 2003 Gulf war. Their study focused on the ﬁrst night of the war on
March 19, 2003, and found that US. news sites relied more heavily on the
military conﬂict, human interest, and media self-coverage frames, while the
responsibility frame was found to be more common among the international sites.
The study concluded that “media are inﬂuenced by the overall political
environment in which they exist,” (p. 35). Another study by Aday, Cluverius, and
Livingston (2005) examined the cable news coverage of toppling of Saddam
Hussein’s statue during the Iraq war in 2003. Their study found that the victory
ﬁame dramatically dominated the amount of battle-related stories. Their
deﬁnition of the victory frame is that the United States has won the war, as
opposed to information that would include alterative frames, such as battle images
that gives the impression that the war continues. In a cross-cultural analysis of
news media coverage of the 2003 Iraq war, Aday, Livingston, and Hebert (2005)
found that although majority of stories on American network and cable television
were objective, they were dominated by episodic battle coverage, “thereby
crowding out other important aspects of the war,” (p. 18). Their study supported

the episodic frames propounded by Iyengar (1991). Their data indicated that

19

twenty-four—hour cable channels—CNN and F ox—used more strategic and
tactical ﬂames when compared to the three traditional broadcast networks or
Arab satellite channel Al Jazeera. The study found that the American television
paid little attention to dissent in the United States or abroad on the Iraq war, and
international diplomacy involving the United Nations. They concluded that the
American viewers got a sanitized version of the war “free of bloodshed, dissent,
and diplomacy, but full of exciting weaponry, splashy graphics, and heroic
soldiers,” (p. 8).

Studies on the Policy Conﬂict Over the Iraq War:

Both, the ﬁrst and second Gulf Wars have been studied in the context of
policy (e.g., Edelrnan, 1993; Entman, 2004; Hiebert, 2003; Luther & Miller, 2005;
Peer & Chestnut, 1995). Edelman’s (1993) work on government policies and
public opinion argued that the 1991 Persian Gulf War was ﬁ'amed as a military
and foreign policy concern by the political elites, which in turn obscured most of
the consequences of the war. The labeling the Gulf War as a military and foreign
policy issue blurred the focus on war causalities and how “the political and
economic beneﬁts that ﬂowed to some from the preparations for war in the Gulf
region,” (p. 233). His study concluded that: “The social world is a kaleidoscope of
potential realities, which can be readily evoked by altering the ways in which
observations are framed and categorized. Classiﬁcations schemes are therefore
central to political maneuver and political persuasion. Typically, they are driven
by the dominant elite’s ideology and prejudice rather than by rigorous analysis or

the aspiration to solve social problems,” (p. 231). According to some scholars

20

(Bennett, 1994; Entman & Page, 1994), when it comes to foreign policy and war
coverage, the government elites, such as the White House, have been shown to
play a privileged role in shaping the news. Bennett and Manheim (1993) analyzed
the ﬁrst Gulf War stories in the New York Times and found that the range of in-
depth views presented was quite narrow. Their study also revealed that media
frames throughout the crisis reﬂected those projected by the administration and its
supporters. They came to a conclusion that the readers lacked reasonable access to
a sufﬁciently wide range of information and opinion to make informed judgments
about the rationale of administration policy. In an analysis of media coverage of
levels of support or criticism accorded to President Bush’s Persian Gulf War
policy in 1991, Peer and Chestnut (1995) found that television news tended to
support the ofﬁcial line more than the newspapers. The thematic nature of
television news forces the broadcast reporters to repeat or cue into what their
sources have said or going to say, whereas newspaper reporters are not
constrained to repeat their sources verbatim.

Entman (2004) examined the coverage of elites’ policy debate on the ﬁrst
Gulf war. He found that the news media coverage of the debate overall appeared
objective, but criticism of the Bush administration was slightly more frequent.
Most of the relevant critical information tended to be displayed less prominently
than support. He argued that much of the criticism was procedural than
substantive. The news media provided basic criticisms of the govemment’s
problem deﬁnition, “which meant discussion of remedies was limited to just the

choice of war or sanctions (Entman, 2004, p. 78).

21

In a preliminary analysis of the government’s handling of the Iraq war,
Hiebert (2003) argues that the use of various techniques of public relations and
propaganda were essential on the part of the US. government to handle the 2003
Iraq war. His study suggests that American government initially framed the Iraq
issue as self-defense story, and later changed the frame to rescue story. The self-
defense frame emphasized Saddam Hussein’s connection to Al Qaeda, terrorism
and weapons of mass destruction. The rescue frame focused on saving the Iraqi
people and bringing democracy to the region. He argued that both these ﬁames
also worked for senior George Bush in the Persian Gulf War (Hiebert, 2003).
While commenting on the government and the role of the media he wrote: “The
government will frame the issues, story line, slogans and catch phrases to serve its
purposes. Only an independent, alert, and responsible press, together with a
concerned citizenry, can force the government to be more honest and forthright,”
(p. 254).

In a study of ﬁaming of the 2003 U.S.—Iraq war demonstrations before and
during the war, Luther and Miller (2005) found the existence of partisan master
frames in texts by pro— and anti—war groups. They argued that frames sponsored
by anti-war groups were more apparent than frames projected by pro-war groups.
The news stories about each group reﬂected the frames of the group in question
more so than the opposing group’s frames. Their pro-war frames included
fighting for freedom and democracy, and threat from weapons of mass
destruction. The anti—war frames were anti—war protest eﬂorts, and global

resistance against the US. ‘empire.’

22

Relevant Iraq War Francs in the 2004 Presidential—Race

The framing of the Iraq war can be studied from an inductive approach
involving analyses of news stories with a view to examine possible frames that
may exist. This approach can point at several possible ways in which an issue
such as the Iraq war can be framed. However, this method can be cumbersome,
often based on small sample, and can be hard to replicate (for a similar argument
see Semetko & Valkenburg, 2000)

Another approach is predeﬁning certain frames as content analytic
variables to examine the extent to which they occur in the news. According to this
approach, it is “necessary to have a clear idea of the kinds of frames likely to be in
the news, because the frames that are not deﬁned a priori may be overlooked,”
(Semetko & Valkenburg, 2000: p. 95). The advantage of this approach is that it
can be replicated, “can cope with large samples, and can detect differences in
framing between media and within media,” (p. 95).

Therefore, under this approach the framing of the Iraq issue can be studied
in several ways, pointed to by past research, such as strategy and issue frames
(Rhee, 1997), episodic and thematic frames (Iyengar, 1991), and human interest,
economic consequence, and morality frames (e.g., Semetko & Valkenburg, 2000).
However, some of these frames are more appropriate than others for a single
policy issue study. The strategy frame approach, for example, is not suitable for
this study as only a single issue is being analyzed in-depth. On the other hand, the
morality frames can be seen as a dimension of a responsibility frame for the

purpose of this study because this research is examining the Iraq issue in the

23

context of election. Speciﬁcally the Democratic challenger could attribute moral
responsibility for the Iraq war to President Bush and his administration. Research
on campaign coverage (Esser & D’Angelo, 2003) also points that proposed
reforms of the McGovern-Fraser Commission resulted in a shift from party-
centered to candidate-centered elections. Candidate-centered campaigns are
organized through the “news media and are characterized by ever-evolving rules
or engagement between candidates and the press corps called media politics,” (p.
618). In these candidate-centered campaigns it is normal to see challengers
framing each other as competent or incompetent in handling of a speciﬁc issue. In
the 2004 presidential race Bush and Kerry framed each other as competent and
incompetent in handling the Iraq issue.

Therefore, this study speciﬁcally investigates the following twelve news
frames that have been identiﬁed in earlier studies that are especially relevant for
this research: ﬁght against terrorism frame, and ﬁghting for freedom and
democracy frame (e.g., Heibert, 2003; Luther & Miller, 2005), Kerry
incompetence frame, and Bush competence frame (e.g., Esser & D’Angelo,
2003), conﬂict frame (e.g., Dimitrova et al., 2005; Reese & Buckalew, 1995),
political ﬁght frame (Patterson & Donsbach, 1996), consensus ﬁ'ame (e.g., Reese
& Buckalew, 1995), policy failure/responsibility frame (e.g., Dimitrova et al.,
2005; Edelrnan, 1993; Entman, 2004; Entman & Rojecki, 1993), economic
consequence frame (Semetko & Valkenburg, 2000), Bush incompetence frame,
and Kerry competence frame (Esser& D’Angelo, 2003), and other frame (frames

not falling into any of the above mentioned categories).

24

These frames are especially suitable to examine news media’s role in
context of the political power play during the 2004 presidential race. According to
previous research (F ico, Clogston, & Pizante, 1988; F ico & Cote, 2002) at higher-
level campaigns challengers have a good chance of getting more coverage than
the incumbent, but the 2004 campaign was different because of the ongoing
conﬂict in Iraq, which meant consistent coverage to the Bush administration that
could have been either positive or negative. As the leader of the nation and
commander in chief of the US. armed forces, President Bush was faced with
election even as the American troops were stationed in Iraq.

The Bush administration was aware of the fact that Democratic Party
would attack his Iraq war policy as several questions remained on the rationale of
going to the war and no clear plan on troop withdrawal. Therefore, President Bush
and his campaign emphasized that the Iraq war was not just an aggression on
another nation, but was in fact a ﬁght against terrorism. President Bush
maintained that the world was better off by removing Saddam Hussein from
power as he had links with terrorist organizations like Al Qaeda. He also asserted
that Iraqi people were liberated from the tyrannical rule of Saddam and that a
stable, democratic government there would serve the end of world peace.

On the other hand, his Democratic challenger, Senator Kerry, left no
opportunity to attack President Bush, holding him responsible for the mess in Iraq
and tying it to a failed Vietnam War policy. Kerry and his supporters kept
attacking President Bush, saying that his administration’s Iraq policy was even

making things worse in Iraq and displacing attention and resources from the War

25

on Terror. Although, the Iraq issue was a point of conﬂict for these two
candidates, an occasional consensus was seen between them, such as their
reactions following the release of Bin Laden tapes threatening elections and
American freedom. They both afﬁrmed that the nation is united against terrorism
and that the nation is not Republican or Democratic when it came to ﬁghting
terrorism.

This study proposes to examine these twelve news media frames in
coverage of the Iraq issue in the context of whether particular frames most
beneﬁted President George Bush or the US. Senator John Kerry. The assumption
is that both political parties were aware of the frames that most beneﬁcially
shaped debate, and actually attempted to place those frames into news coverage.
Obviously too, however, journalists are themselves gatekeepers and, as such, their
decisions to give access to sources and their frames matter in the policy debate.
Moreover, the journalists themselves are frame providers, and those frames may
themselves beneﬁt one or the other side.

Based on the above rationale and previous studies (Dimitrova et al., 2005 ;
Esser & D’Angelo, 2003; Hiebert, 2003; Luther & Miller, 2005; Patterson &
Donsbach, 1996; Reese & Buckalew, 1995; Semetko & Valkenburg, 2000)
following frames are developed for this study:

Pro-Bush Frames:

Fight Against Terrorism Frame: The focus is on equating Iraq war with
ﬁght against terrorism. Bush administration’s emphasis that the world is better off

with the ouster of Saddam Hussein.

26

Fighting for Freedom and Democracy Frame: The focus is on bringing
democracy to the world in general and to Iraq in particular.

Kerry Incompetence Frame: The focus is on Kerry being a ﬂip-ﬂopper and
unﬁt to lead the nation in its ﬁght against terrorism and the war in Iraq.

Bush Competence Frame: The focus is on Bush being capable of leading
the nation successfully against its ﬁght on terrorism and the war in Iraq.

Neutral Frames:

Conﬂict Frame: The focus is on claims by Bush or Kerry that they would
make American safer from Islamic terrorists, or win the war in Iraq.

Political Fight Frame: Here sources explicitly identify that one or both
sides, Republicans and Democrats, are using the Iraq issue to secure some
political or electoral gain, as a tactical strategy.

Consensus Frame: The focus is on Bush and Kerry agreeing on some
aspect of the Iraq issue. For instance, they agreed that Bin Laden tape released
during campaign will not deter them from their ﬁght against terrorism.

Pro-Kerry Frames:

Policy Failure/Responsibility Frame: The focus is that the Iraq issue is a
policy failure that squanders international goodwill. It emphasizes that the Bush
administration is responsible for the Iraq war and its problems including military
difﬁculties, civilian casualties and prisoner abuse. Here, the focus is on Bush
administration rather than President Bush as an individual.

Economic Consequence Frame: The focus is on economic burden being

caused at the domestic front due to the war in Iraq.

27

Bush Incompetence Frame: The focus is on Bush as a headstrong and
arrogant individual who is unﬁt to lead the nation in its ﬁght against terrorism and
the war in Iraq. Here, the focus is on President Bush ad an individual.

Kerry Competence Frame: The focus is that Kerry will keep focus on
terrorism, not on Iraq. Kerry will be smarter than Bush. Kerry will get allies, and
world support to hunt down and kill terrorists.

Overall, then, the rationale of this study is that President Bush and his
supporters framed the Iraq issue as war against terrorism and that only the Bush
administration could ensure a secure and safe America. On other hand,
Democratic Senator Kerry and his supporters framed the Iraq issue as a failure
attributable to the Bush administration that only a change in administration could
correct. Assessing the news media frames promoting these views therefore
illuminates the power play between political parties reﬂected in the mass media.
This study will also help understand whether the news media tilted coverage to
favor the frames of one or the other side, or provided other frames that were more
neutral.

Factors InﬂuencingNews MedmcLIssueLrames

Shoemaker and Reese (1996) suggest that various factors intrinsic and
extrinsic to the media organizations inﬂuence mass media content (also see, Gans,
1979; Tuchman, 1978). They argue that from an organizational routines approach,
media content is inﬂuenced by the way journalists and their news companies
organize work (e.g., Johnson, Braima, & Sothirajah, 2000; Peer &. Chestnut,

1995). This study builds on this hierarchical model, and examines the inﬂuence of

28

intrinsic media factors such as nature of the medium (cable vs. network channels),
and race and gender of journalists, and an extrinsic media factor, political and
non-political sources, on news media framing of the Iraq issue. Speciﬁcally, this
study contributes to knowledge in this area by developing a model that takes
account of medium, journalist, and source inﬂuence on framing in a policy
conﬂict context.

Certainly there will be a difference between the way network news
channels and twenty-four hour cable news channels organize their work and
thereby differently inﬂuence their media content. In other words, there will be
differences based on the priorities and goals of the news channel managers who
set priorities, allocate resources and establish acceptable news processing
routines. With regard to the inﬂuence of individual journalists on the media
content, Shoemaker and Reese (1996) argue that media’s practices and routines
will suppress such effects on the content. However, recent research (e.g., Craft &
Wanta, 2004; Fico & Freedman, 2001; Freedman & Fico, 2005; Gidengil &
Everitt, 2003; Sutcliffe, Lee, & Soderlund, 2005; Zeldes & Fico, 2005) suggests
inﬂuence of journalists race and gender on the news content. Finally, Shoemaker
and Reese (1996) point at several factors extrinsic to the media organizations that
inﬂuence the media content, and prominent among these kinds of factors are
sources. Sources can be political, economic and social elites, interest groups, and
ordinary people.

The theoretical premise of this study is that cable and broadcast network

news may differ in their ﬁaming of news because of the differences in the amount

29

of time they can devote to stories and in the interest of their targeted audiences in
news. For instance, Bae (2000) found that due to time differentiation there were
signiﬁcant differences in the percentage of unique stories between the cable and
broadcast network newscasts. Other literature suggests that the Fox News cable
channel has a conservative bias while a network channel like CBS has a liberal
bias (e.g., Aday, Livingston, & Hebert, 2005; Katovsky & Carlson, 2003;
McChesney, 2002). While comparing the difference between broadcast and print
medium coverage of Persian Gulf War policy debate, Peer and Chestnut (1995)
found that television news tends to support the government more than the
newspapers, and attributed this feature to the structural makeup of the respective
media.

On the inﬂuence of journalists’ demographics on news frames, it has been
argued that the race and gender of journalists will have minimum effect on the
news content (Weaver & Wilhoit, 1986). However, some studies (Aday,
Livingston, & Hebert, 2005; Armstrong, Wood, & Nelson, 2006; Rodgers &
Thorson, 2003; Nonis et al., 2003; Zelizer & Allan, 2002) posit that the nature of
American journalism shifted after September 11 in ways that affect the inﬂuence
of journalists’ race and gender on news frames (e.g., Gidengil & Everitt, 2003;
Sutcliffe, Lee, & Soderlund, 2005; Freedman & Fico, 2005). In a study of female
news professionals in local and national broadcast news during the buildup of the
Iraq War, Armstrong et al. (2006), for example, found gender to be more a key
factor in the national news content than in local news. They argue that post-

September 11th world is dominated by stories of war, globalization, foreign

30

policy, and the opportunity of reporting at national level does not look promising
for female journalists. This is evident by the fact that male reporters and anchors
were shown often in the coverage of the Iraq war, and that less than 10% of
journalists embedded with troops were women (Armstrong et al., 2006). Rodgers
and Thorson (2003), in a study of print news professionals, based on a gender
model, suggested that differences in socialization might be the reason for
difference in the news coverage by male and female print journalists. However,
they also qualify this, asserting, “Newsroom norms and practices may operate as a
conformin mechanism that socializes all reporters, regardless of gender, to
maintain the male-dominated power structure of larger news organizations,”
(p.673).

Finally, news sources have been always a point of interest for mass media
scholars examining how news content is inﬂuenced by political and social elites,
who are often used as sources by the journalists (e.g., Allen et al., 1994; Entman,
1989). Such sources are given more or less scope in stories both to set the issue
agenda and to frame particular issue. In a policy conﬂict, the access given to
contending sources to do this relates to fairness and balance in journalism.

Research relating to these types of inﬂuence is discussed more thoroughly
below.

Nature of the Medium:

Mass communication and political literature suggests that the nature of the
medium inﬂuences media content including frames (Benson, 2004; D’Alessio &

Allen, 2000; Johnson, Braima, & Sothirajah, 2000; Peer & Chestnut, 1995;

31

Semetko & Valkenburg, 2000; Shoemaker & Reese, 1996; Wicks & Walker,
1993). Some scholars argue that studies of the nature of medium are usually
limited to one kind of medium such as newspapers or television (D’ Alessio &
Allen, 2000). In a meta-analysis of media bias they argued that “[news media] are
perceived as a monolithic block. Critics discuss ‘the media’ ...as a unitary whole
and not as a set of not only different companies but even drastically different
media,” (p. 137). Therefore, their study concluded that media analysis should be a
‘gross-level examination,’ (p. 137). In another study, Benson (2004) argues that
more studies are required to understand the inﬂuence of structural characteristics
of media systems to the production of journalistic discourse about politics.

There are some studies which indicated the inﬂuence of nature of the
medium on media content. For example, Aday, Livingston, & Hebert, (2005) in
their analysis of coverage of the Iraq war observed that twenty-four—hour cable
channels use more strategic and tactical frames when compared to traditional
broadcast networks. This may be because cable channels have longer newscasts
and can devote more time to speciﬁc issues or events. Iyengar (1991) suggested
that television news tends to use more episodic frames compared to print medium.
In a content study of framing of European politics, it was found that sober and
serious newspapers and television news programs more often used responsibility
and conﬂict frames, whereas sensationalist outlets often used the human interest
frame (Semetko & Valkenburg, 2000). A product differentiation study of cable

and broadcast networks (Bae, 2000) found that cable newscasts covered

32

proportionally more government, politics, human interest, feature stories, whereas
broadcast networks reported more health and welfare stories.

A few studies found the Fox News Channel coverage of Iraq war to be
more jingoistic and pro-administration when compared to other cable channels
and networks (Aday, Livingston, & Hebert, 2005; Katovsky & Carlson, 2003).
Citing a survey, Aday, Livingston, & Hebert, (2005) wrote: “FNC (Fox News
Channel) was founded in response to what many conservatives in America felt to
be a liberal bias in the mainstream media and currently attracts an audience more
heavily made up of Republicans than other networks’ audiences,” (p. 6). In
another study the Project for Excellence in Journalism (2002) examined news
coverage soon after the September 11 attacks and found Fox News Channel to be
more favorable to the administration than other networks. However, the Center
for Media and Public Affairs (CMPA, 2003) studied the coverage of the Iraq war
and found that Fox News Channel did not carry substantively more positive
stories than the three broadcast networks (Aday, Livingston, & Hebert, 2005).

Other studies suggest that while Fox News Channel may have a
conservative bias, a network channel like CBS has a liberal bias (e.g., Aday,
Livingston, & Hebert, 2005; D’ Alessio & Allen, 2000; Katovsky & Carlson,
2003; Lowry & Shidler, 1995, 1998; McChesney, 2002). Speciﬁcally, in a meta-
analysis of media studies of presidential elections, D’ Alessio and Allen (2000)
found that television network news had a pro-Democratic Party bias. Their data
indicated that about 52.7 percent of airtime went to Democrats, while Republicans

got roughly 47.3 percent of airtime. Based on these ﬁnding they reasoned: “This

33

is not a large difference, although it is larger than some would like,” (p. 149).
While emphasizing the importance of presidential campaigns, D’Alessio and
Allen (2000), argued that ideally unbiased media should cover Republican and
Democrat sides equally. They pointed out that “the winners achieve enormous
power for themselves and their parties. . .The potential consequences of media
bias, particularly if it can be shown to have inﬂuenced the campaign (as is so
often the claim), are therefore similarly enormous,” (p. 138). In a study of 1992
presidential campaign, Lowry and Shilder (1995) discovered that broadcast
networks were fair in giving equal time to Republican and Democratic candidates,
but the overall sound bites of noncandidates revealed a liberal news bias. In
another longitudinal study of 1992 and 1996 presidential campaigns, Lowry and
Shidler (1998), while emphasizing the importance of analyzing sound bites by
noncandidates and candidates in broadcast network news, argued that
“Republicans tend to be negatively bitten more often than Democrats,” (p. 277).

Bae (2000) in his study emphasized that among cable channels, Fox News
Channel and MSNBC started out to be different from CNN. Fox News Channel is
considered to have a conservative slant because of the political personalities of
Rupert Murdoch and Roger Ailes. MSNBC promoted itself as futuristic,
technology-oriented cable channel by linking its television news program with the
Web site. It also attempted to differentiate from CNN both in news coverage and
target audience (Bae, 2000).

An earlier Stempel (1988) study comparing news on CNN primetime, PBS

(MacNeil Lehrer News Hour), and three traditional evening network newscasts on

34

ABC, CBS, and NBC, concluded that broadcast newscasts, CNN, and PBS did
not offer the same news. On 14 topic categories, the study found some differences
along with moderate correlation between CNN and three broadcast networks on
the news topics covered. The study also found that CNN heavily emphasized
popular amusements, was less dependent on news from New York and
Washington DC. and reported more international stories.

The Media Institute (1983) in a comparison of newscast on broadcast
networks and CNN found that CNN ’5 business and economic news stories were
more balanced and less sensational than those of the broadcast networks. CNN ’3
reporting, however, was of less depth compared to network reporting, despite
two-hour length of Primenews. There was no signiﬁcant difference in the news
priority of the networks and CNN. While examining a relationship between
different broadcast types, Johnson et al., (2000) found that the different media
reaches different audience and that the media are more powerful at the beginning
of the campaign. In another study of media coverage of President Bush’s Persian
Gulf War policy in 1991, Peer and Chestnut (1995) found that television news
tended to support the ofﬁcial line more than the newspapers. The thematic nature
of television news forces the broadcast reporters to repeat or cue into what their
sources have said or going to say, whereas newspaper reporters are not
constrained to repeat their sources verbatim. In another study examining the
difference in the live coverage of the Persian Gulf War in CNN and other
broadcast networks, Wicks and Walker (1993) found that CNN ’5 coverage was

more event-oriented, while the other three networks provided more analysis.

35

Race and Gender of Journalist:

Before discussing the inﬂuence of race and gender of journalists on media
frames, it will be helpful to focus on journalists themselves serving as sources.
Framing literature in political communication and mass communication
emphasizes the role of journalists in presenting news on various issues and events
(Callaghan & Schnell, 2001; Edelman, 1993; Entman, 1991, 1993; Gamson,
1992; Gitlin, 1980; Iyengar, 1991; Luther & Miller, 2005; Pan & Kosicki, 1993;
Valkenburg et al., 1999). These scholars assume that framing is inevitable in
journalism and that journalists intentionally or unintentionally frame their
representations of issues or events in order to make them accessible to a mass
audience. Scholars have long argued that journalists have to narrate a story within
a limited time or space. They employ certain frames to simplify and give meaning
to the issues and events and to sustain audience interest (Ball-Rokeach &
Rokeach, 1987; Jasperson et al., 1998; Valkenburg et al., 1999).

Even though journalists can use myriad ways to frame the news, the
literature seems to suggest that there are at least four ways in which news is
generally framed: conﬂict frame, human interest frame, responsibility frame, and
economic consequence frame (Semetko & Valkenburg, 2000; Valkenburg et al.,
1999). According to some scholars (Callaghan & Schnell, 2001; D’Angelo, 2002),
journalists play the dual role of serving as conduits for transmitting such frames
sponsored by the political elites, and also select and simplify (i.e., frame) certain
aspects of the stories in order to present them in interesting and comprehendible

form to their audience. In a framing study of the gun control issue, for instance,

36

Callaghan and Schnell (2001) found that both politicians and interest groups
employed several interpretive issue frames and keenly contested to put their
preferred themes on the agenda. At times, however, journalists intervened in their
framing process, especially as the debate matured.

With the growing emphasis on newsroom diversity and source diversity
(Kerschbaumer, 2003; Monroe, 2003) this study explores whether the race and
gender of journalists will have inﬂuence on framing of the Iraq issue. Although,
Weaver and Wilhoit (1986) argue that demographics such as the race and gender
of journalists will have minimum effect on the news content, recent mass
communication research suggests other possibilities (e.g., Armstrong et al., 2006;
Craft & Wanta, 2004; Freedman & Fico, 2005; Gidengil & Everitt, 2003;
Richardson & Lancendorfer, 2004; Sutcliffe, Lee, & Soderlund, 2005; Zeldes &
Fico, 2005). In a study combining content analysis and experiment, Gidengil and
Everitt (2003) analyzed the television news coverage of women candidates in
1993 and 1997 Canadian elections. Based on gendered mediation thesis, they
argued that conventional news ﬁames construct politics in stereotypically
masculine terms. While cautioning that the gender differences between female
and male journalists should not be overstated, their study concluded that “female
journalists may be more likely than male journalists to react critically to assertive
and nonstereotypical behavior on the part of female leaders. It is also possible that
the journalists’ choice of language reﬂects the constraints of the gendered context
in which they work,” (p. 225). Quoting Ross and Srebemy (2000, p. 94) they

wrote, “women journalists often resort to even more macho reporting styles than

37

their male colleagues as if to prove their professional mettle.” Their study also
found that female reporters are more likely to use female sources, and male
reporters have a tendency to use more male sources, which is similar to ﬁndings
of other studies (e.g., Zeldes & F ico, 2005).

Craft and Wanta (2004) examined the inﬂuence of female editors and
reporters on the news agenda and found that newspapers with female editors often
had less negative content than those with male editors. A study by Rodgers and
Thorson (2003) revealed that though male and female journalists often carry out
reporting assignments differently, organizational variables including the
circulation size and the ratio of male-to-female editors, mediated those
differences. As argued by some scholars (Armstrong et al., 2006; Rodgers and
Thorson 2003), the gender model suggests that males and females have different
workplace roles and behaviors because they are socialized differently.

In an analysis of the US. network television news coverage of political
crises in the Caribbean, Sutcliffe et al., (2005), discovered subtle differences in
the way male and female reporters ﬁamed stories. Although, there was no
dramatic difference between male and female reporters’ coverage of Caribbean
political crises, the study did indicate that female reporters had a tendency to
include more stories on social and economic concerns. Even though there was a
broad consistency in male and female reporting, the study found that female
reporters and anchors were slightly more likely to use female sources in their
sources. In a content study of race and gender of reporters in the 2000 presidential

campaign, Zeldes and Fico (2005) found that presence of women and minority

38

reporters led to more diverse source usage. Women and minority reporters were
more likely to use and give time to women and minority noncandidate sources
compared to male and white reporters. Women and minority reporters also gave
more story time and attention to sources such as experts and ordinary people.
Another study by Freedman and Fico (2005) found that female reporters have a
greater tendency than their male colleagues to use female sources. “At
newspapers where there is a ‘critical mass’ of minority journalists willing to argue
for or against particular slants or ﬁames being used with stories, the character of
those stories should differ from those in newspaper with a less substantial
minority presence,” (Gandy, Kopp, Hands, Frazer, & Phillips, 1997, p. 165).

However, van Zoonen (1994) argues that it is not always clear that just by
adding women to media organizations will transform media output. According to
van Zoonen, there is no monolithic women’s viewpoint. “Because all knowledge
is socially situated, however, and because women’s situations differ from those of
men, the possibility exists that their reporting will differ from that of men,” (p.
149). With regard to female news reporters, agenda setting suggests that an under-
representation of women, and any resulting biases in the content of reporting, may
feed back into both public perceptions and government policy (Sutcliffe et al.,
2005). For instance, the overall number of women appearing as reporters and
sources on the Iraq issue during presidential campaign, at some level, may
translate into American public opinion on that particular issue.

Some scholars argue the possibility of male bias in news reporting

affecting the content of that news story (e.g., Mills, 1997; Sutcliffe et al., 2005).

39

These scholars emphasize that the choice of news stories, how they are reported,
and selection of sources, may be affected by those who are reporting the story. To
underscore this point, Gallagher (2001) argues that “the power to deﬁne public
and media agendas is mame a male privilege,” (p.3). She asserts that media
content “reﬂects a masculine vision of the world and of what is important,” (p. 7).
A few scholars argue that women are underrepresented as journalists and this may
create distortions in news reporting. Further, this underrepresentation of women
may result in certain news stories, or attributes of news stories, being neglected
(e.g., Sutcliffe et al., 2005).

Partisan Balance of Stories:

Previous studies have shown that media frames are inﬂuenced by sources
(e.g., Allen et al., 1994; Callaghan & Schnell, 2001; Entman, & Page, 1994; Fico
& Cote, 2002; Luther & Miller, 2001; Page, Shapiro, & Dempsey, 1987;
Parmelee, 2002; Ryan, Carragee, & Meinhofer, 2001). According to these studies,
sources can be broadly categorized into partisan political elites such as
Republicans and Democrats, and nonpartisan sources such as experts, journalists
and ordinary people (e.g., Carter, Fico, & McCabe, 2002; Fico & Balog, 2003;
Fico & Freedman, 2001; Freedman & Fico, 2004, 2005). Some studies have
emphasized that partisan elite sources and journalists control the media frames
(e.g., Callaghan & Schnell, 2002; Fico & Cote, 2002; Fico & Freedman, 2001;
Valkenburg et al., 1999). Therefore, this study argues that sources who get more
prominence and attention in turn get more opportunity to set a story’s frame. For

instance, in a study of newspaper coverage of controversial issues, Fico and

40

Sofﬁn (1995) found that anti-war sources opposing the president’s Persian Gulf
war policy dominated the news, whereas major public opinion polls conducted
during the study’s sample period indicated about 70 percent support to the war.
The ﬁndings of this study can be attributed to news media’s power to select
sources and frame stories. Another study by F ico and Freedman (2001) assessed
the newspaper coverage of 1998 govemor’s race in Michigan and discovered that
assertions of the candidates and their partisan supporters dominated the story
leads compared to experts or other sources. They also found, however, that about
5 percent of the stories had leads containing reporters’ own interpretations or
conclusions that were not substantiated in the story, which indicates journalists’
independent ability to inﬂuence the news content. Freedman and Fico (2004)
assessed the use of expert sources in newspapers coverage of gubernatorial
elections and found that more prominently displayed stories cited more experts
than other stories, as did stories that were more balanced between electoral
opponents. They argued that experts’ observations could provide insights that
might inﬂuence voters. Another study by Fico and Balog (2003) on local conﬂict
reporting by Michigan newspapers revealed that though experts and nonpartisan
sources appeared in almost half of stories, partisan opponents on the issues got
more prominent and extensive coverage.

In a study of news media framing of collective action, for example, Ryan,
Carragee, and Meinholder (2001) found that journalistic frames are affected by
the frames sponsored by several social actors such as political elites and

advocates. They argue that news stories become a platform for ﬁarning contests in

41

which these actors compete in promoting their deﬁnitions of political issues. In
line with earlier studies (Entman & Page, 1994; Luther & Miller, 2005; Donohue,
Tichenor, & Olien, 1995), they concluded that ﬂaming contests favor political and
economic elites.

Some studies analyzed the difference between the media frames sponsored
by the partisan sources such as Republicans and Democrats (e.g., Fico & Cote,
2002; Jacoby, 2000; Parmelee, 2002). In an analysis of 2000 presidential primary
Videocassettes Parmelee (2002) found that Democrats sponsored values frame,
and leadership and electability frame, whereas Republicans projected reclaim
Regan ’s legacy frame, and change frame. In a media bias study of 1998
govemor’s race in Michigan, Fico and Cote (2002) discovered that Republican
John Engler emphasized an education issue agenda, whereas his Democratic
challenger, Geoffrey F ieger, indulging in insulting campaign comments
personally attacking Engler, managed instead to ﬂame himself as an issue. J acoby
(2000) examined the framing of government spending and found that
“Republicans focus on broad, general appeals (e.g., ‘government spending must
be cut!’), while Democrats aim at more speciﬁc forms of programmatic outlays
(e.g., ‘It is important to ﬂrnd medical care for the elderly!’), (p. 750). Citing the
works of Sears (1993), Allen et al. (1994) argued that conservative Republicans
usually discuss an issue or event in terms of ‘welfare,’ while a liberal Democrat
may discuss the identical issue or event as ‘helping the needy.’ Morreale (1991)
conducted a textual ﬁ'ame analysis of Ronald Reagan’s 1984 presidential

campaign ﬁlm, A New Beginning, which was aired on television nationwide. She

42

argued that the Republicans used ideological, mythic, and rebirth frames in the
ﬁlm to create a ‘reality.’ In a content study of the meta-coverage of 2000
presidential campaign Esser and D’Angelo (2003) found that Republican
candidates, Bush and Cheney used a strategy public frame to deal with their
negative coverage about their pasts.

Although the above mentioned studies have emphasized that political
elites and journalists dominate the media with their ﬂames; it is also true that
American news media reﬂect frames supplied by others, including experts and
ordinary citizens. For example, stories may focus on an individual, or emphasize
human emotions (human interest frame) (McLeod & Hertog, 1992; Valkenburg et
al., 1999).

In sum, this study argues that partisan sources such as Republicans and
Democrats, and nonpartisan sources such as experts, ordinary people and
journalists will inﬂuence the news media framing of the Iraq issue. However, not
all such sources will be equal in their power to do this, and will certainly be
different in the frames they advance in an intense and consequential policy

conﬂict.

43

CHAPTER 3: Linkages of Frames to Sources

Research Questions and Hypotheses

The goal of this study is to examine framing of the Iraq issue in television
news. The idea is to study the extent to which fight against terrorism, ﬁghting for
freedom and democracy, Kerry incompetence, Bush competence, conﬂict,
political ﬁght, consensus, policy failure/responsibility, economic consequence,
bush incompetence, Kerry competence, and other frames were found in the news
framing of the Iraq issue during election coverage between Labor Day and
election day. These frames are worth exploring because the Bush administration
was trying to ﬂame the Iraq issue as ﬁght against terrorism, ﬁghting for freedom
and democracy (e.g., Hiebert, 2003; Luther & Miller, 2005), Kerry is incompetent
to deal with the Iraq issue, and that Bush is competent to lead the nation in ﬁght
against terrorism and solving the Iraq issue. On the other hand, Kerry was trying
to attack his Republican opponent by framing the Iraq issue as policy failure of
Bush administration, that it was responsible for creating the whole Iraq issue (e.g.,
Iyengar, 1991; Entman, 1991), which is causing huge economic burden on the
US. domestic front, and that Bush is incompetent to lead the nation against war
on terrorism, and Kerry is a better leader to solve the mess in Iraq. The news
media is known to make stories interesting and easy to understand by highlighting
the conﬂict element in the story, especially during presidential race they tend to
pitch each campaign story as intense political ﬁght, and, however, when some

accord is reached between the contending sides journalists underscore the

44

consensus aspect (e.g., Reese & Buckalew, 1995; Semetko & Valkeburg, 2000;
Valkenburg et al., 1999).

There was a considerable political debate and sharp divide in public
opinion over the question of war in Iraq. Therefore, the Iraq issue got increased
news media coverage. The objective of this study is to gain more insight into
national news coverage framing by examining journalistic news production
processes in a political conﬂict context (see Gamson, 1992; Neuman et al., 1992;
Callaghan & Schnell, 2001). Such a perspective enriches the understanding of
news frames and acknowledges that frames are the result of interactions between
journalists, sources, and the broader political and media system context (Pan &
Kosicki, 1993; de Vreese et al., 2001)

In line with previous research (Esser & D’Angelo, 2003; de Vreese et al.,
2001; Semetko & Valkenburg, 2000) this study ﬁrst explores how the Iraq issue
was framed in the news media. Therefore, ﬁrst research question is:

RQl: What kinds of frames emerged most frequently in the coverage of

the Iraq issue during the 2004 presidential campaign?

Another theoretical interest is to compare the use of ﬂames between
network and cable news and to consider whether there are important differences
among different news outlets. Previous studies (e.g., Bae, 2000; Semetko &
Valkenburg, 2000) have argued that there are signiﬁcant differences between
cable and broadcast networks in the way they cover issues and events. CNN ’5
reporting, for example, was of less depth compared to network reporting (Media

Institute, 1983; Wicks & Walker, 1993). Also, Stempel (1988) concluded that

45

CNN and broadcast networks did not offer the same news. Bae (2000) suggested
that cable news would differ from network news as the former is of longer
duration. Thus, based on the same rationale it can be assumed that cable and
network channels will differ in their framing of the Iraq issue. Therefore,
following hypothesis is posited:

H1: Network and cable newscasts differed in the framing of the Iraq issue.

The literature (Aday, Livingston, & Hebert, 2005; Bae, 2000; Center for
Media and Public Affairs, 2003; Project for Excellence in Journalism, 2002) has
argued that Fox News Channel with a conservative slant will be more pro-
administration when Republicans are in power. These studies emphasize that Fox
News Coverage will be different from other cable and network channels. Also, the
literature suggests that while the Fox News cable channel has a conservative bias
(e.g., Aday, Livingston, & Hebert, 2005), and a network channel like CBS has a
liberal bias (e.g., Katovsky & Carlson, 2003; McChesney, 2002).

Therefore, on the basis of this rationale it can be argued that Fox News
Channel will use more pro-administration ﬂames such as ﬁght against terrorism,
and ﬁghting for freedom and democracy while covering the Iraq issue. And CBS
will use more anti-administration frames such as responsibility, and policy failure.
Thus, this study posits following hypotheses:

H2: Fox News Channel used more pro-Bush ﬂames compared to

other cable and network channels.
H3: CBS News used more pro-Kerry frames compared to other cable

and network channels.

46

Previous studies have suggested that journalists play up conﬂict element in
the story to make it appealing for audience (e.g., Semetko & Valkeburg, 2000;
Valkenburg et al., 1999). This argument allows to ask following research question
and hypothesis:

RQ2: What kinds of frames were most frequently used by journalists to

cover the Iraq issue?

H4: Journalists most often used the political ﬁght ﬂame over other frames

while covering the Iraq issue.

To assess the inﬂuence of race and gender of journalists on frames (e.g.,
Richardson & Lancendorfer, 2004; Sutcliffe et al., 2005) the following research
question is posed:

RQ3: Did race and gender of a journalist affect framing?

Speciﬁcally, Gedengil and Everitt (2003) argue that female journalists are
likely to use more aggressive verbs while reporting on certain events such as
speech of female leaders. Sutcliffe et al. (2005) suggested that there are subtle
differences in the news frames used by male and female reporters. Therefore, the
following hypothesis is posited:

H5: Women journalists were more likely to use pro-Kerry frames while

covering the Iraq issue.

Literature points at the inﬂuence of partisan balance of stories on news
media content (e.g., Fico & Soffm, 1995; Fico & Cote, 2002; Jacoby, 2000;

Parmelee, 2002). Thus, following hypotheses are proposed:

47

H6: Stories with greater Republican partisan balance used more pro-Bush

ﬂames.

H7: Stories with greater Democratic partisan balance used more

pro-Kerry frames.

RQ4: Were Republican or Democratic sourced ﬂames more likely to be

found in the coverage?

Several studies have suggested the inﬂuence of experts and nonpartisan
sources on news content (e.g., Carter, Fico, & McCabe, 2002; Fico & Balog,
2003; Freedman & Fico, 2004, 2005).

RQS: What kind of ﬂames were most ﬂequently used by expert and

Nonpartisan sources to describe the Iraq issue?

48

r- ~ I. ',,.
”liqu- .' .l

u
I

’ .nl' " 'r
(MW ”tin rW’

'.r

 

 

A Model of Inﬂuences on Framing

Overall then, this study explores the following model:

 

Nature of Medium
(Network vs. Cable)

 

 

 

 

i

 

Nature of Journalist
(Race & Gender)

 

 

H1, H2, & H3

 

 

 

 

H4. H5. R02. &R03

 

 

 

 

 

 

 

 

Iraq Issue Frames
RQI

 

 

 

Nature of Source
(Rep, Dem, & N-part.)

 

 

/ H6, H7, RQ4, &RQ5

 

 

 

 

Figure 1: A Model of Inﬂuences on Framing: Model explicating the inﬂuence of
communication variables on the Iraq issue ﬂaming. Dotted boxes include
hypotheses and research questions of this study.

Following the hierarchical model set out by Shoemaker and Reese (1996) ,

this model assumes that news organization leaders are key in recruiting and

shaping the conduct of media journalists, and that those journalists themselves are

key gate keepers for news sources and the frames they set in the news. A content

study by itself cannot test the linkages by which managers select and train

reporters, or the considerations reporters bring to their selection of sources. A

content study can, however, examine differences in ﬂames and their ﬂequencies

49

 

that are associated with differences in medium (networks vs. cable), journalist
characteristics (male vs. female, and white vs. non-white) and source type
(political partisans, journalists and other non-partisan sources). More pointedly,
this study assesses whether these differences in medium, journalist characteristics
and source type have predictive power in identifying ﬂames used or not used in
this issue.

This study can also assess the relative inﬂuences of these factors on
ﬂaming. If these factors demonstrate framing inﬂuence in this empirical study,
they can be retained in larger, more elaborate and reﬁned models. Empirical
testing is necessary, of course, because however plausible and “obvious” a

relationship might be, life is frequently more complicated, especially political life.

50

CHAPTER 4: Content Analyzing Iraq War Frames

Method

In this digital age there are more avenues for getting news, but for most
Americans, television news is still the most common source of information about
developments in the country (Bae, 2000; PEJ 2006). A recent study on state of the
American news media by the Project for Excellence in Journalism revealed that
more than 27 million people still rely on network television for news. Another
study by the Pew research center (2004a) suggests that people are still getting
their presidential campaign news ﬂom cable channels. For instance, cable
channels such as FOX and CNN have registered a 4 percentage point growth
(ﬂom 34% to 38%) as source for campaign news. Therefore, this study carried out
a quantitative content analysis of evening newscast of three networks ABC, CBS,
and NBC; and the three cable-TV channels, CNN, MSNBC, and FOX.

Sim

The period of study was between Labor Day (September 6, 2004) and the
day of election (November 2, 2004). According to the literature (e.g., Esser &
D’Angelo, 2003), this time period during the presidential race is important
because presidential candidates, political parties, interest groups, media, and
voters tend to focus more on crucial issues a few months before the election day.
For the three network news channels, ABC, CBS, and NBC, all the evening
newscasts between 6:30 and 7 pm. were videotaped for this time period.
Similarly, for cable-TV—CNN evening newscast by Wolf Blitzer, MSNBC

evening newscast by Keith Olbermann, and FOX evening newscast by Shepard

51

Smith — were videotaped for the time period. These cable newscasts were chosen
because they included stories of daylong events and in comparison came close to
the evening newscasts of the traditional broadcast networks.

The Iraq issue coverage in the context of the presidential campaign was
signaled by explicit station graphics and verbal transitions by anchors. Daily Iraq
issue coverage could consist of stories running in isolation or grouped into
election segments. The content analysis emphasized stories containing assertions
from Bush or Kerry partisans, reﬂecting the campaigns’ direct attempts to sway
voters, ﬂom non-partisans, and ﬂom journalists themselves.

Codingjrocedure

A two-member team of coders was trained and supervised at a large
Midwestern university. The unit of analysis was stories, coded for the number of
seconds devoted to total time and time on the Iraq issue. A news story on Iraq
issue is a story that covers Iraq issue in the context of 2004 US. presidential
election. Iraq issue (in the election context) can be any story involving search for
weapons of mass destruction in Iraq, conﬂict between the US. forces and
insurgents/terrorists in Iraq, efforts to reconstruct Iraq and bringing democracy. to
the region, government facing criticism on its Iraq policy and other similar news
stories related to Iraq. Stories totally on war events, however, were not coded
absent an explicit reference to the 2004 election.

To carry out the quantitative content analysis of television news a coding
protocol was developed. Questions were resolved at regular meetings. After

repeated pretesting of the codebook, Iraq news stories were coded.

52

Appendix I includes the ﬁnal code sheet and protocol for the study.

Measures:

For the purpose of this study, framing deﬁnitions are drawn from the
works of Entman (1993), Gitlin (1980), and Gamson and Modigliani (1987).
Moreover, Iraq war stories were observed for ﬂames that were different or more
ambiguous than those suggested by the ﬂaming literature.

Framing is deﬁned as the way sources or journalists describe or present
the Iraq issue. Sources (partisan and nonpartisan) or journalists will select some
aspect of the Iraq issue and make it more salient in the news story, “in such a way
as to promote a particular problem deﬁnition, causal interpretation, moral
evaluation, and/or treatment recommendation,” (Entman, 1993, p. 52). The Iraq
issue frames in general were identiﬁed by assessing what the sources emphasized
about the issue in stories. In this study, journalists are considered as the ﬂamers of
the story only when there are no sources or verbs of attribution such as said, felt,
and hoped linked to sources in the story.

Dependent Variable: Iraq Issue Frames

For operationalization of the Iraq issue ﬂames, this study largely drew
ﬂom the measures developed by following studies: Dimitrova et al., (2005), Esser
& D’Angelo (2003), Hiebert (2003), Luther & Miller (2005), Reese & Buckalew
(1995), and Semetko and Valkenburg (2000). Each news story was examined for
the ﬂames projected by all sources including Republicans, Democrats, non-
partisans, and journalists. It is possible that each source may be projecting

multiple ﬂames. Frames projected by each source category were timed. While all

53

the frames were timed and recorded, the ﬂame getting the maximum airtime in
seconds for each source category was considered the dominant ﬂame. Speciﬁc
ﬂames were decided by the kind of assertions given by the source in the story.

A preliminary assessment of stories were made to illuminate relevant
ﬂames and their approximate frequency. In addition, observed ﬂames were
categorized as favoring Bush, Kerry or essentially neutral based on how they
deﬁned problems, attributed causes, identiﬁed those morally responsible, and
pointed to possible solutions (e.g., Entman, 1994). Consequently, ﬂame emphases
were determined by the following deﬁnitions below:

Pro-Bush ﬂames include:

Fighting against terrorism frame: Here the focus is on equating the Iraq
war/issue with the ﬁght against terrorism. The source (s) will explicitly emphasize
that the Iraq war/issue is a ﬁght against terrorism. For example, President Bush
may emphasize that Iraq was axis an axis of evil nation under Saddam Hussein’s
regime and that world is better off without him being in power. Example of key
words or phrases to help identify this ﬂame are ﬁght against terrorism, axis of
evil, world better off without Saddam, supporting terrorists, weapons of mass
destruction, and protecting America ﬂom terrorists.

Fighting for ﬂeedom and democracy ﬂame: Here the focus is on equating
the Iraq war/issue with bringing ﬂeedom and democracy to the Iraqi people. The
source (s) will explicitly emphasize that the Iraqi people have been liberated ﬂom
the dictatorship of Saddam Hussein. Efforts are being made to establish a

democratically elected government in Iraq. Example of key words of phrases to

54

help identify this ﬂame are democracy, liberation, ﬂeedom, rescuing Iraqi people
ﬂom tyranny of Saddam, and bringing ﬂeedom and democracy to Iraqi people.

Kerry incompetence ﬂame: Here the focus is that Kerry is a ﬂip-ﬂopper.
Source (5) would ﬂame Kerry as indecisive and that he changes his issue position
on the Iraq war, which makes him unﬁt to lead the nation, speciﬁcally on
handling the Iraq issue. Example of key words of phrases to help identify this
ﬂame are ﬂip-ﬂopper, indecisive, and Kerry does not make a good leader, etc.

Bush competence ﬂame: Here the focus is that Bush is a better
commander-in-chief to lead the nation in right direction. Bush can do a better job
than Kerry at the domestic ﬂont as well as in foreign policy such as the war in
Iraq. Example of key words of phrases to help identify this ﬂame are I will be a
better leader, I will be a better commander-in-chief, and I can lead the nation in
right direction, etc.

Neutral ﬂames include:

Conﬂict ﬂame: Here the focus is on Bush and Kerry claims in the context
of the terrorism and Iraq issue that they would provide a safer America. To be
considered a conﬂict story, sources should be explicitly referring to the broader
conﬂict rather than to the election. The Republican and Democrat partisans will
argue that they are committed to the ﬁght to protect America. Example of key
words or phrases to help identify this ﬂame are I will make a better leader, I will
handle Iraq situation better, and attack etc.

Political ﬁght ﬂame: Here sources explicitly identify that one or both

sides, Republicans and Democrats, are using the Iraq issue to secure some

55

political or electoral gain, as a tactical move to win domestic electoral support.
Example of key words or phrases to help identify this ﬂame are shoring up
support, wedge issue, political gain on support, winning voter strategy, Kerry
attacked Bush, Bush attacked Kerry, and the presidential campaign is in a
statistical dead-heat etc.

Consensus frame: Here the focus is on Bush and Kerry agreeing or
expressing consensus on some aspect of the Iraq issue. For instance, when the Bin
Laden tapes were released, Bush and Kerry agreed with each other that they will
never give in to the demands of terrorists. In such situations Bush and Kerry
emphasized that America remains united, and is not Republican or Democratic
when it comes to tackling terrorism. Example of key words or phrases to help
identify this ﬂame are we are united to ﬁght against terrorism, we will not give in
to the threats of terrorists, and we are united and not Republican or Democrat
when it comes to ﬁghting terrorism etc.

Pro-Kerry ﬂames include:

Policy failure/responsibility ﬂame: Here the focus is that the Iraq issue
reﬂects policy failure of the Bush administration. That the Bush administration is
responsible for the Iraq issue. For instance, Kerry might argue that the Bush
administration took a hasty decision of invading Iraq and it reﬂects a poor
thinking and plamring on the part of current government Kerry might allege that
President Bush misled the country by claiming that Iraq stored weapons of mass
destruction and took the nation to the war without taking the citizens into

conﬁdence. Example of some key words and phrases to help identify this ﬂame

56

are policy failure, poor planning, hasty decision, responsible, misled, reason, and
lied

Economic consequence ﬂame: Here the focus is on economic
consequence of the Iraq war. For example, Kerry might point out that the ill-
planned Iraq war has cost over $200 billion, which could have been used for
education and generating jobs. Example of some key words and phrases to help
identify this ﬂame are economic cost of the Iraq war, billions of dollars have been
wasted on the war, war is causing economic burden on the country etc.

Bush incompetence ﬂame: Here the focus is on Bush’s incompetence to
handle the Iraq issue. The focus is more on Bush as an individual. For instance,
Kerry might allege that President Bush misled the country by claiming that Iraq
stored weapons of mass destruction and took the nation to the war without taking
the citizens into conﬁdence. Example of some key words and phrases to help
identify this ﬂame are cannot handle the Iraq issue, misled, arrogant, and poor
planning, etc.

Kerry competence ﬂame: Frame projected here is that Kerry will keep
focus on terrorism, not on Iraq. Kerry will be smarter than Bush. Kerry will get
allies and world support to hunt down and kill terrorists. Example of some key
words and phrases to help identify this ﬂame are competence, allies, winning
international supporters, global test etc.

Other ﬂame: Use the other ﬂame when the frames projected by the
sources don’t fall into any of the deﬁned categories above, or when the projected

ﬂame (s) could not be clearly established.

57

Appendix II contains example passages for each of these frame measures
taken ﬂom sample stories.

Reliabilig and Validity

A two-person coder. reliability test was conducted on about 10% of stories
relevant for this study randomly sampled ﬂom the population of stories.
Percentage of agreement for all the ﬂame identiﬁcation variables ranged ﬂom 89
% to 98%. Scott’s Pi computations to correct for chance agreement ranged ﬂom
.86 to .97. Those variables measured on ratio-level scales (seconds devoted to
ﬂames) were assessed for coder reliability using Pearson correlation coefﬁcients,
which ranged ﬂom .96 to 1.0 (Riffe, Lacy, & Fico, 2005).

Measures used in this study to examine ﬂames were based on past
research, theory, and the logical assumption that the news media’s ﬂaming of the
issues depends upon various factors such as medium, the journalists themselves,
and sources. Unfortunately, as Damore (2005) argues, it is difﬁcult to ﬁnd a
statistical method to easily assess the validity of framing data. Therefore,
alternative criteria should be used.

One appropriate and straightforward validity assessment is face validity.
Face validity is established when the deﬁnitions of concepts used in a study have
a straightforward correspondence to observed indicators of the concept. The
ﬂame deﬁnitions used in this study rely on words with manifest and denotative
reference to concepts such as competence, costs, failure and democracy. In
particular, then, the arguments partisans use to advance their causes rely on words

whose meanings are common and explicit. Frames based on those words therefore

58

share the explicitness of the meanings of component words and are relatively less
abstract or ambiguous compared to other concepts (e.g., news story “quality” or
“bias”).

Other research has established the predictive validity of ﬂaming.
Predictive validity can be established with reference to experimental ﬂaming
research that links ﬂame concepts to those established in the minds of those
exposed to the ﬂames. For instance, a study by Valkenburg, Semetko, and de
Vreese (1999), revealed issue-speciﬁc ﬂaming effects. In an experiment they
presented readers with two news stories, one crime and the other on economy,
ﬂamed in terms of conﬂict, human interest, attribution of responsibility, and
economic consequences. Their results revealed that the way in which news stories
ﬂamed the issue had a signiﬁcant effect on readers’ thoughts on both issues.
Respondents who only read the story with the human interest ﬂame emphasized
more emotions and individual implications in their responses. Those who read
stories with the conﬂict ﬂame more often expressed thoughts that involved
conﬂict. And the ones who read stories with the economic consequence ﬂame
thought more in terms of costs and ﬁnancial implications (V alkenburg et al.,
1999)

Another study by de Vreese (2004), showed that television news ﬂames
“have the ability to direct viewers’ thoughts when conceiving of a contemporary
political issue,” (p. 45). His study found that respondents who watched a news
story ﬂamed in terms of conﬂict or economic consequences expressed thoughts

about the issue that was in line with how the news was ﬂamed.

59

Independent Va_r_iable Measures

Independent variables in the study, pointed to by theory and past research,
are the nature of the medium, the nature of journalists, and the nature of sources.

The nature of the medium for this study is broadcast network versus cable
news programs. The broadcast networks include ABC, CBS, and NBC. The
networks channels have been covering presidential campaigns roughly for more
than seven decades and are known to uphold the traditional journalistic values
without being very sensationalistic in their approach to cover the news. On the
other hand, the cable channels like CNN, Fox News Channel, and MSNBC are
comparatively a recent phenomenon. Structurally, cable news channels are
different ﬂom the traditional broadcast news networks as cable channels operate
on a twenty-four hour cycle. The cable news channels have more newshole to ﬁll
and were basically launched to provide something more and different, in terms of
news coverage, to their audience.

Literature has pointed out that journalists not only serve merely as
conduits for information, but also inﬂuence the information as gatekeepers. They
have the power of selecting news and making certain attributes more salient than
others. In other words, journalists can put their own spin on the news that they
cover. In this study, the journalists’ role as story sources was examined and an
assessment made about whether their race and gender in any way inﬂuence the
ﬂaming of media content. It is possible that the journalists can themselves
sponsor any of the ﬂames being examined in this study through their own

assertions (i.e., assertions not explicitly attributed to identiﬁed sources).

60

The third independent variable in this study is source. For the purpose of
this study, sources are categorized as partisan sources (Republicans, Democrats)
and, non-partisan sources (anyone who, doesn’t explicitly assert a Republican or
Democrat leaning), and journalists themselves (when they do not attribute
information to any source). A partisan source may be the candidate, staff, family
or party supporters, or simply a citizen who supports one of the candidates for
president. A single partisan assertion establishes the partisanship of the source for
all election coverage. A source who has not made a partisan assertion in any story
is considered a non-partisan source. Assertions lacking any attribution to a source
are considered the journalist’s.

Two other variables were constructed for this study because of their
potential inﬂuence on viewer exposure to candidate ﬂames. One variable assessed
the dominant source in the story. The other assessed the story’s overall partisan
balance (i.e., did the story “favor” Bush or Kerry).

The source in each of the four independent variable categories who gets
maximum airtime in a story was considered to have set the dominant ﬂame in that
story. For example, in a story running 45 seconds, if a Republican source get 20
seconds for the ﬂame, a Democratic source got 15 seconds to set the ﬂame, a
non-partisan source got 5 seconds, and the news reporter got another 5 seconds
for the ﬂame, then in this case the dominant ﬂame is being set by a Republican,
which is most likely to be one of the ﬂames supporting the president’s reelection.
Speciﬁc pre-defmed ﬂames in this study will be linked to speciﬁc sources by the

story emphasis given by the sources.

61

Finally, the partisan balance of each story was assessed following
procedures used in a series of broadcast studies by (e.g., Carter, F ico, & McCabe,
2002; Fico, Zeldes, & Diddi, 2004). These procedures assess which candidate
obtained more story prominence based on position in the story, total air time for
assertions, presence in visuals and on-air quotation. Assessing story partisan
balance this way therefore illuminates ﬂames given more story prominence and,
potentially, ﬂames getting more audience attention.

Data Analysis

All election-related Iraq stories aired between Labor Day and the day of
election by the six stations were analyzed for the presence of ﬂames. In other
words, the study data come ﬂom the universe of coverage, not ﬂom a random
sample. The unit of analysis here was the story and story time devoted to the Iraq
issue. In each story, ﬂames projected by the Republicans, Democrats, experts,
ordinary citizens and journalists were identiﬁed and timed in seconds. The source
getting the maximum airtime for her or his ﬂame was considered to be setting the
dominant story ﬂame, but multiple ﬂames could be coded for each story.

Data were analyzed ﬂom three perspectives. First, entire stories were
assessed for the presence of the dependent variable ﬂames that were dominant in
the stories. Second, total seconds were calculated for the Iraq war stories
(including multiple ﬂames) and the kinds of sources setting these ﬂames. Finally,
the ﬂaming seconds themselves were calculated for each ﬂame by each source in

each story.

62

This study especially compares ﬂame ﬂequencies in both total story
seconds and in ﬂaming seconds. Total story seconds devoted to ﬂames is
necessary to assess because ﬂames may be more or less important based on a
story’s total time and the proportion of the time devoted to ﬂaming. In other
words, total story seconds relates to the probability of a viewer’s exposure to any
ﬂames. Attention to ﬂaming seconds, on the other hand, permits an assessment of
who was most favored in ﬂame attention compared to the opponent, an
assessment related to journalistic fairness and balance.

Rules for Assessing Research Ouestiom Hypotheses

The assessment of important differences in ﬂaming was made in three
ways. The ﬁrst looked at the partisan balance of pro-Bush and pro-Kerry ﬂames
in stories, in total seconds and in ﬂaming seconds. Obviously this relates to the
potential inﬂuence of imbalance on public opinion, as well as the journalistic
norm of fairness and balance. Second, particular attention was paid to important
differences in use of speciﬁc ﬂames. In particular, did different ﬂames emerge
most ﬂequently based on medium, journalist characteristics or source type?
Finally, neutral ﬂame usage was assessed in order to illuminate differences
attributable to differences in medium, journalists, and types of source. In other
words, for example, were ‘non-partisan’ sources actually neutral or balanced in
their ﬂaming or were they facilitating (deliberately or inadvertently) a partisan
agenda?

In studies relying on randomly sampled data, identiﬁcation of

generalizable differences is made by tests of statistical signiﬁcance. However, that

63

is a less useful procedure for this study which analyzes the universe of data for a
particular population and time period. Therefore, it is necessary, absent
appropriate statistical tests, to identify how a hypothesis will be judged to be
supported or not, or observed differences to be considered signiﬁcant or not.

Consequently, this study sets an apriori rather than a strictly statistical
standard for hypotheses testing and the assessment of research question ﬁndings.
Given that most of the data for this study can be assessed in proportions, a
difference of at least 10 percentage points is considered to be “important.” For
example, if male journalists use the pro-Bush, Bush competence ﬂame in 35% of
their stories and female journalists use it in 45% of theirs, that will be considered
an important difference in ﬂame usage attributable to differences in gender.

The 10 percentage point criterion draws ﬂom the logic of statistical
inference. A difference of proportions test in a random sample of about 350 cases
would require a 10 percentage point difference for statistical signiﬁcance. The
present study, based on much larger numbers of stories and story seconds,
therefore sets a higher bar for the identiﬁcation of signiﬁcant and important
differences.

Moreover, for hypothesis testing, a hypothesis is considered fully
supported only when all of the relevant assessments discussed above equal or
exceed 10 percentage points. For example, a fully supported hypothesis that
network and cable channels differ in their ﬂaming behaviors would require at
least a 10 percentage point difference in their partisan imbalance (whether

favoring Bush or Kerry), in their use of at least one speciﬁc ﬂame, and in their

use of one or more of the neutral ﬂames. Differences less than 10 percentage
points in all of those assessments would signal complete non-support for a
hypothesis. Differences equaling or exceeding 10 percentage points in some but
not all relevant assessments would be considered partial support.

Rule Application to Research OucsMnd Hypotheses

Applying these rules, the ﬁrst Research Question probing the kinds of
ﬂames emerging most ﬂequently in the coverage of the Iraq issue was answered
by comparing the percentage of time devoted to different ﬂames. This was done
by identifying the ﬂame dominant in a story and by calculating multi-ﬂame usage
in total story seconds and total ﬂaming seconds.

Research Question 2 probing the kinds of ﬂames ﬂequently set by
journalists was answered by computing the percentage of time journalists
themselves devoted to different ﬂames. Research Question 3 exploring the impact
of race and gender of journalists was answered by computing partisan imbalance,
differences in use of each ﬂame and differences in use of the neutral ﬂames. Each
of these three assessments was done to compare white and non-white journalists,
and male and female journalists.

Hypothesis 1 through 3 positing differences in the ﬂaming of the Iraq
issue between the broadcast networks and cable newscasts was answered by
computing the proportions of the ﬂames used by each outlet. Cable and network
differences were then assessed by calculating the relative overall attention in

ﬂaming balance between pro-Bush and pro-Kerry ﬂames, by assessing

65

differences in particular ﬂames and by comparing the network and cable
channels’ use of the neutral ﬂames.

Hypothesis 4 positing that journalists would primarily use political ﬁght
ﬂame to cover the Iraq issue was answered by computing percentage of time
devoted to different substantive ﬂames and to the neutral ﬂames.

Hypotheses 5 through 7 focusing on types of ﬂaming sources and
Research Questions 4 and 5 were answered in similar way by computing relevant
proportions. Republican and Democratic sources are explicitly assessed on the
kind of substantive ﬂames they used and how they differed in the ﬂames

emphasized. It is assumed that these sources are not likely to use neutral ﬂames.

Assessing Relative Strength of Inﬂuences on Framing

 

Assessing the relative strength of inﬂuence of medium, journalistic
characteristics and source type on ﬂaming is made in three ways. First, these
inﬂuences are assessed by the degree to which relevant hypotheses were
supported. A second assessment is made by considering the types of sources
setting ﬂames most ﬂequently in the total ﬂaming seconds of all the stories.

However, an important political and journalistic question concerns the
relative degree to which these factors produced imbalance in the use of pro-Bush
or pro-Kerry ﬂames in the news. This relates to the normative standard that news
media should be “fair and balanced,” as well as to possible charges of bias.
Imbalance also relates, of course, to the partial shaping of public opinion through

the partisan ﬂarrring (whether Republican or Democrat) of an issue.

66

Consequently, a regression analysis was also run in which story was the
unit of analysis and story partisan ﬂaming imbalance the dependent variable. This
ratio level dependent variable was obtained by subtracting total pro-Kerry framing
seconds in a story ﬂom total pro-Bush ﬂaming seconds, and then taking the
absolute value of that result. The resulting scale, therefore, reﬂects imbalance in
election ﬂaming, regardless of the candidate favored, and could run ﬂom 0 (Bush
and Kerry seconds exactly equaled one another) to some positive number higher
than 0 (depending on the size of the difference in Bush and Kerry ﬂaming
seconds).

Independent variables in the regression were medium type (cable vs.
network), journalist race (white vs. non-white), journalist gender (male vs.
female), and source type (partisan vs. all others). These variables were coded so
that the dichotomous values were given scores of 0 or 1.

Regression Beta weights were then used to assess the relative strength and
direction of these inﬂuences. Given the way the speciﬁc variables were assigned
values, a positive Beta would indicate that imbalances were produced by cable
channels, by non-white reporters, by women reporters or by partisan ﬂame
sources. Beta sizes (whether positive or negative) illuminate the relative strength

or importance of the variable on framing imbalance in stories.

67

CHAPTER 5: Who Framed The War and How

Results

Description of Dia by the Number of Stories

In all, 445 Iraq issue news stories were analyzed for this study involving
three networks, ABC, CBS, and NBC, and three cable channels, CNN, FOX, and
MSNBC, during 2004 presidential campaign between Labor Day and the day of
election. Table 1 summarizes number of stories by each television channel.

The Fox News Channel (FNC) produced largest number of stories with
24%, while MSNBC, which lagged signiﬁcantly behind FOX, CNN, and NBC,
produced the lowest with 8% (see Table 1),. The three networks together
produced about 47% of the stories, while the three cable channels together

produced nearly 52% of Iraq stories.

Table 1: Percentage of the Iraq Issue Stories in Television Channels.

 

 

 

 

 

News channels Percentage of stories
All networks 47%
ABC 17%
CBS 11%
NBC 19%
All cable channels 52%
CNN 20%
FOX 24%
MSNBC 8%
Total 100%
(445)

 

 

 

(The sum of percentages may not exactly equal 100 due to rounding. Number in
parenthesis indicates total number of stories).

68

The majority of the stories, about 52%, were reporter packages followed
by anchor lead-ins with 22%, anchor vo/sots with 10%, and the remaining were
other story types (see Table 2). Certainly the differences among these ways of
reporting news is dramatic, with reporters carrying the major burden of reporting
the Iraq election stories. All other anchor-related material still amounted to 47%
of the total, but nearly half of that is made up of the anchor lead-ins, many of
which merely introduced reporter packages and lacked substantive campaign

information.

Table 2: Percentages of Types of Stories.

 

 

 

 

 

 

 

 

 

 

Story Type Percentage
Reporter package 52%
Anchor lead-in 22%
Anchor vo/sot 10%
Anchor in-house interview 7%
Anchor voice over 6%
Anchor reader 2%
Reporter live 2%
Total 100%

(445)

 

 

 

(Number in parenthesis represent number of stories. The sum of percentages may
not exactly equal 100 due to rounding).

The data indicate that networks and cable channels differ in important
ways in how they cover stories. About 59% of network stories, but only 44% of

cable ones, relied on reporter packages (see Table 3). Network anchor lead-ins

69

reported about 30% of their Iraq stories compared to 14% for cable channels. The
cable channels differed ﬂom the networks by using more of anchor vo/sot with
17% to 3%, anchor in-house interview with 12% to 2%. This reﬂects a structural
difference in news production and reporting between networks and cable
channels, with cable channels placing more emphasis on their anchors, especially
for interviewing news sources in studios or by live feeds. Although both networks
and cable channels may attempt to “brand” their news offerings by their choice of
anchors, cable anchors are not only more visible but also much more active in the

on-air news process (for similar argument see Armstrong et al., 2006, p. 81).

Table 3: Percentages of Story Types by Networks and Cable Channels.

 

 

 

 

 

 

 

 

 

 

 

 

Story type Networks Cable channels
% %
Reporter package 59 44
Anchor lead-in 30 14
Anchor vo/ sot 3 17
Reporter live 3 1
Anchor reader 2 2
Anchor in-house interview 2 12
Anchor voice over 1 10
Total 100% 100%
(21 l) (234)

 

 

(The sum of percentages may not exactly equal 100 due to rounding. Numbers in
parenthesis represent number of stories.)

About 89% of the stories were reported by white journalists and 11% by

non-whites (see Table 4). The data in Table 4 also reveal that 16% of the Iraq

70

issue stories were covered by female journalists and 84% by male reporters.
These proportions are consistent with a number of other studies (e.g., Armstrong
et al., 2006; Zeldes and Fico, 2005) that show overwhehningly male and white
presence among on-air reporters. Indeed almost every on-air anchor for these
news programs was a white male during the period of this study. It was possible
to see a female and non-white anchor for these news programs, but usually during

weekend newscasts.

Table 4: Race and Gender of Journalists and Story Percentages.

 

Race Gender

 

White Non-white Male Female

 

 

Story percentages 89% 11% 84% 16%

 

 

 

 

 

 

N = 445

When it came to using non-partisan sources, whether they set the
dominant ﬂame or not, MSNBC accounted for the highest proportion of them in
its stories (32%) (see Table 5). NBC, ABC, and CNN used non-partisan sources
almost half the time of that of MSNBC. FOX had least number of stories with
non-partisan sources (3%). It should be noted, however, how few Iraq stories
incorporated such sources during the 2004 election. In other words, the Iraq
debate was carried on mostly by those whose vested interest was in winning the

election.

71

Table 5: Number of Stories with Non-partisan Sources.

 

MSNBC NBC ABC CNN CBS FOX Total

 

 

Stories 32% 18% 18% 16% 13% 3% 100%
138)

 

 

 

 

 

 

 

 

 

(The sum of percentages may not exactly equal 100 due to rounding. Numbers in
parenthesis represent stories with non-partisan sources).

Description of Dat_a by the Time Devoted to Stories and Frames

The overall seconds total for the 445 stories was 24,539 seconds,
including 16,759 (68%) ﬂaming seconds (see Table 6). CNN had largest share of
both total story seconds (29%) and ﬂaming seconds (31%). CBS had the least of
both, 10% story seconds and 9% ﬂaming seconds. Indeed, the correlation between

story seconds and ﬂaming seconds is nearly perfect.

Table 6: Percentages of Story and Framing Seconds.

 

 

 

 

 

 

 

 

 

News channels Story seconds Framing seconds
CNN 29% 31%

NBC 19% 18%

FOX 16% 19%

ABC 14% 14%
MSNBC 11% 10%

CBS 10% 9%

Total 100% 100%

(24,539) (16,759)

 

 

 

 

(The sum of percentages may not exactly equal 100 due to rounding. Numbers in
parenthesis represent total number of story seconds and ﬂaming seconds).

72

 

For all the story types, the percentages of story seconds is also similar to
the percentages of ﬂaming seconds (see Table 7). The reporter package, with
more than 60% of all seconds, was the dominant story type used by the television
news, while the anchor reader with 1% was used the least. Anchor in-house
interviews were another ﬂequently used technique for the reporting the Iraq issue,
followed by anchor vo/sot stories. When story seconds are considered, it is clear
that anchor lead-ins played a minor role in substantive reporting on Iraq.

Table 7: Type of Story and Percentage of Time.

 

 

 

 

 

 

 

 

 

Story Type Story seconds Framing seconds
Reporter package 62% 60%
Anchor in-house interview 18% 18%
Anchor vo/ sot 9% 10%
Anchor lead-in 6% 7%
Anchor voice over 2% 2%
Reporter live 2% 2%
Anchor reader 1% 1%

 

 

 

 

(Numbers in parenthesis represent total story length and ﬂaming seconds).

With regard to just ﬂaming seconds on the Iraq issue, 87% were reported
by white journalists and 13% by non-whites (see table 8). About 20% of ﬂaming
seconds on the Iraq issue stories were covered by women journalists and 80% by
male reporters. An interesting observation is that non-white and women reporters

had a slightly higher proportion of ﬂaming seconds compared to their

73

representation in total stories (see Table 4). However, white and male journalists
were still overwhelmingly more likely to report stories in which the Iraq issue

obtained one or more of the ﬂames measured in this study.

Table 8: Race and Gender of Journalists and Framing Seconds Percentages .

 

Race Gender

 

White Non-white Male Female

 

 

Frannng seconds 87% 13% 80% 20%

 

 

 

 

 

 

N = 16,759

Non-partisan sources (non-journalists who made no statements explicitly
supporting a candidate) accounted for 1,125 seconds in just 38 stories. MSNBC,
with 53%, got most of these non-partisan ﬂaming seconds (see Table 9). Fox with
2% got the least proportion of non-partisan ﬂaming seconds. Data in Table 9 also
indicate that ABC devoted the least time to the non-partisan sources (1%), even
though they were included in 18% of its stories. These differences are partly
accounted for by the news formats used in network and cable channels. MSNBC,

for instance, used long in-house interviews conducted by the anchor.

Table 9: Non-partisan Framing Seconds and Stories.

 

Non-partisan MSNBC NBC ABC CNN CBS FOX Total

 

Framing seconds 53% 16% 1% 26% 3% 2% 100%
(1,125)

 

 

Stories 32% 18% 18% 16% 13% 3% 100%
(38

 

 

 

 

 

 

 

 

 

(The sum of percentages may not exactly equal 100 due to rounding. Numbers in
parenthesis represent ﬂaming seconds and number of stories).

74

Reseagh Questions_ and Hypotheses

Framing Patterns in Coverage

The ﬁrst Research Question probed the kinds of ﬂames found most
frequently in the television news coverage of the Iraq issue during 2004
presidential campaign.

While multiple ﬂames were recorded for most of the stories, a dominant
ﬂame was determined for each story based on the simple majority of time that a
particular ﬂame received. An analysis of data to determine a dominant ﬂame for
each of the 445 stories revealed that pro-Kerry ﬂames were dominant in 41%, 10
percentage points more than dominant pro-Bush ﬂames (see Table 10). Moreover,
dominant partisan frames far exceeded neutral ones, hardly a surprising ﬁnding in
election coverage. Interestingly, few Iraq stories lacked any of the ﬂames
identiﬁed for this research, and only a few more Iraq stories contained dominant
ﬂames not deﬁned for this study.

Among stories with dominant ﬂames, the Kerry incompetence ﬂame was
the most ﬂequent pro-Bush ﬂame (see Table 10). The policy failure/responsibility
was the most ﬂequent pro-Kerry ﬂame, followed by the Bush incompetence
ﬂame. All other dominant partisan ﬂames were included in less than 10% of
stories.

Among dominant neutral ﬂames, the political ﬁght ﬂame was clearly the
most ﬂequent. Almost no dominant neutral ﬂames emphasized the national
conﬂict in the war on terror, or points of consensus or agreement among

candidates on waging that war.

75

Table 10: Dominant Frames in Stories.

 

 

 

 

 

 

 

Frames Erequency 0f .
ames 1n stones
All Pro-Bush 31% (140)
Fighting against terrorism 4%
Fighting for ﬂeedom/dem 2%
Kerry incompetence 20%
Bush competence 5%
All Neutral 21 %(93)
Conﬂict 1%
Political ﬁght 20%
Consensus 0%
All Pro-Kerry 41%(176)
Policy failure/ resp. 23%
Economic cons. 2%
Bush incomp. 12%
Kerry competence 4%
Other 5% (24)
No ﬂame 3% (12)
Total 100%(445)

 

 

 

 

(The sum of percentages may not exactly equal 100 due to rounding. Numbers in
parenthesis represent absolute numbers.)

Analyzing all the ﬂames in stories revealed that pro-Kerry ﬂames were
present in 70%, 2 percentage points more than pro-Bush ﬂames (see Table 11).
Partisan ﬂames exceed neutral ones, but by a much smaller margin than when
dominant ﬂames were considered.

The Kerry incompetence ﬂame was the most ﬂequent pro-Bush ﬂame
(38%). The policy failure/responsibility ﬂame was the most frequent pro-Kerry
ﬂame (34%), followed by Bush incompetence ﬂame (22%). All other partisan
ﬂames were included in less than 15% of stories (see Table 11).

Among the neutral ﬂames, the political ﬁght ﬂame was clearly the most

ﬂequent (57%) in that category and among all ﬂames overall. Moreover,

76

whatever substantive partisan ﬂames a story might contain, they were frequently
presented along with a political ﬁght ﬂame emphasizing how the issue helped or
hurt the candidate’s electoral campaign.

Again, the other neutral ﬂames, emphasizing the common conﬂict of the
US. with terrorists or points of agreement of Bush and Kerry, were nearly absent
in coverage (see Table 11). This may reﬂect an acrimonious, highly partisan
electoral battle which entirely focused on divisions. It may also reﬂect journalistic

decisions that points of agreement between the two candidates were just not

newsworthy.

Table 11: Frames in News Stories.

 

 

 

 

 

 

 

Frequency of
Frames ﬂames in stories
All Pro-Bush 68% (297)
Fighting against terrorism 11%
Fighting for ﬂeedom/dem. 5%
Kerry incompetence 38%
Bush competence 14%
All Neutral 59% (264)
Conﬂict 1%
Political ﬁght 57%
Consensus 1%
All Pro-Kerry 70% (315)
Policy failure/ resp. 34%
Economic cons. 5%
Bush incomp. 22%
Kerry competence 9%
Other 10% (43)
Total (445)

 

 

 

(The sum of percentages exceed 100 due to the presence of multiple ﬂames in
stories. Numbers in parenthesis represent absolute numbers.)

77

Total story seconds devoted to ﬂames in stories also is necessary to assess
because ﬂames may be more or less important based on a story’s total time and
the proportion of time devoted to ﬂames. In other words, total story seconds relate
to the probability of a viewer’s exposure to any ﬂames.

Considering total story seconds, pro-Bush and pro-Kerry ﬂames amounted
to just under 60% of the all story seconds (see ﬁrst column in Table 12). About a
third of total story seconds were used for non-ﬂaming purposes such as
introducing reporter packages or providing accounts of campaign activities.
Overall, then, viewers had a good chance of encountering some type of ﬂame set
by the Bush and Kerry partisans, although far less than a certain one.

Viewers had a better chance of encountering a Kerry ﬂame than a Bush
one, but not a much better chance. In total story seconds, pro-Kerry ﬂames
received the most seconds (29%) in the coverage of the Iraq issue, closely
followed by pro-Bush seconds (25%) and neutral seconds (13%) (see Table 12).
No-ﬂame seconds made up 30% of the total.

With regard to individual ﬂames in total seconds, the pro-Kerry policy
failure/responsibility received the most seconds (16%), closely followed by pro-
Bush Kerry incompetence (15%) and neutral political ﬁght (11%) ﬂaming
seconds (see Table 12). The Bush incompetence ﬂame received about 8% of total
seconds, whereas all the remaining ﬂaming seconds were less than 10%. Overall,
then, even the three most ﬂequently used partisan ﬂames made up less than 40%

of the total coverage, and the two most ﬂequently used ﬂames made up less than

78

a third of the total coverage. All other partisan ﬂames could only rarely be

encountered.

Table 12: Percentages of Total Story Seconds and Framing Seconds.

 

 

 

 

 

 

 

 

Total story Framing
Frames seconds seconds
All Pro-Bush 25% (5,925) 36% (5,925)
Fighting against terrorism 3% 5%
Fighting for ﬂeedom/dem. 2% 3%
Kerry incompetence 15% 21%
Bush competence 5% 7%
All Neutral 13% (2,991) 19% (2,991)
Conﬂict 1% 1%
Political ﬁght 11% 17%
Consensus 1% 1%
All Pro-Kerry 29% (7,064) 42% (7,064)
Policy failure/ resp. 16% 23%
Economic cons. 2% 2%
Bush incomp. 8% 12%
Kerry competence 3% 5%
Other 3% (779) 5% (779)
No-ﬂame 30% (7,780) 0% (0)
Total 100% 100%
(24,539) (16,759)

 

 

 

 

(The sum of percentages in columns may not exactly equal 100 due to rounding.
Numbers in parenthesis represent total number of story seconds and ﬂaming
seconds.)

Attention to ﬂaming seconds, permits a better assessment of who was
most favored in ﬂame attention compared to the opponent.

Considering the ﬂaming seconds on the Iraq issue, pro-Kerry frames with
42% of ﬂaming seconds were emphasized more than pro-Bush ﬂaming seconds,
which made up 36% of the ﬂaming coverage, as indicated by the data in the last

column of Table 12. With regard to individual ﬂames measured, as reported in the

79

last column of Table 12, the pro-Kerry policy failure/responsibility ﬂame (23%,)
closely followed by pro-Bush Kerry incompetence (21%) and the neutral political
ﬁght ﬂames (17%) were most ﬂequent in the overall coverage of the Iraq issue in
television news. The pro-Kerry Bush incompetence (12%) and the pro-Bush Bush
competence (7%) ﬂames were also found to some extent in the coverage, whereas
remaining ﬂames obtained less than 5% of seconds each in the ﬂaming seconds
coverage. When just ﬂaming seconds are considered Kerry’s advantage increases
to 6 percentage points compared to 4 percentage point advantage in total story
seconds. Again, by the standards used in this study, the Kerry advantage is not
considered to be signiﬁcant.

This ﬁnding of a consistent Kerry advantage in stories, total seconds and
in framing seconds, however, is consistent with research showing that the news
media tend to give more coverage to a challenger during higher-level elections
(e.g., Fico, Clogston, & Pizante, 1988; Fico & Cote, 2002). An explanation for
challenger getting more time in news media coverage can be attributed to the
objective norms of journalistic practices. In other words, Bush being the president
would get media coverage while doing other activities such as calling on people
affected by hurricane Jeanne in Florida amidst his presidential campaign in
September 2004. On the other hand, challenger Kerry would only get coverage
when he does something newsworthy such as addressing a public meeting.
Therefore, news media might make a conscious effort to give more coverage to
the challenger in order to bring him on par with the incumbent in terms of media

exposure.

80

In the 2004 presidential campaign, these data suggest that Democrats, led
by Senator Kerry, were successful in making Iraq a contentious issue by framing
it as a failed policy of the Bush administration, claiming that an arrogant and
shortsighted President Bush was incompetent to resolve the Iraq issue.

For instance, quoting Senator Kerry, NBC reported on September 20,
2004: “The president's policy in Iraq has not strengthened our national security; it
has weakened it... We have traded a dictator for chaos that has left America less
secure.” An ABC report on October 18, 2004, cited Kerry saying, “Despite the
president's arrogant boasting that he's done everything right in Iraq and that he's
made no mistakes, the truth is beginning to come out, and it's beginning to catch
up with him.”

However, the data also show that the Bush camp countered by ﬂaming
Kerry as a ﬂip-ﬂopper who kept changing his positions on resolving the Iraq
issue, claiming that a leader who sends mixed messages cannot be a good
commander in chief during testing times like the crisis in Iraq. For example, a
CNN report on September 23, 2004, quoted President Bush’s criticism of Kerry:
“I believe a leader must be consistent and clear and not change positions when
times get tough. And the times have been hard. These are hard times. But I
understand what mixed messages do. You can embolden an enemy by sending
mixed messages.” Another report by ABC on October 18, 2004 quoted Bush:
“Having gone back and forth so many times, the Senator ﬂom Massachusetts
[Kerry] has now ﬂip-ﬂopped his way to a dangerous position. My opponent, my

opponent ﬁnally has settled on a strategy, a strategy of retreat.”

81

 

 

 

0i

 

is

It

56’6‘

Positive Republican ﬂaming portrayed Bush as determined to crush
terrorism and stay on course till the Iraq issue was resolved, thereby making him a
better leader. For instance, in an ABC report on October 18, 2004, Bush was
quoted saying: “And in this time of choosing, I want all Americans to know you
can count on me to ﬁght our enemies and defend our freedom.”

Journalists played up the Iraq issue positions of President Bush and
Senator Kerry and primarily ﬂamed it as political ﬁght between the two, a pattern
also found in past research (e.g., Semetko & Valkenburg, 2000; Valkenburg et al.,
1999). For example, Tom Brokaw introduced a campaign story with the following
lead: “President Bush and John Kerry are at war over the war in Iraq. And while
the president ties it to terrorism, the issue that serves him best, Senator Kerry
opened a new ﬂont today linking the war to the American economy and social
issues,” (NBC, September 8, 2004).

Some stories did not fall into the apriori ﬂames deﬁned in this story and
were, therefore, coded as “other ﬂame.” An example of the other ﬂame is the
following news story by ABC’s reporter Terry Moran: “(Off Camera) In a
statement, Bremer says he believes current troop levels are sufﬁcient, and that his
opinion that more were needed related only to the early stages of the occupation.
He also says he strongly supports President Bush's reelection...” (ABC World
News Tonight, October 5, 2004).

A close look at the data for individual channels’ ﬂaming of the Iraq issue
reveal signiﬁcant differences among them. CBS News devoted 42% ﬂaming

seconds to pro-Kerry frames versus 25% to pro-Bush ﬂames, a signiﬁcant

82

difference of 17 percentage points in favor of Kerry (see data in bold in Table
13). These ﬁndings are consistent with previous studies, which attribute liberal
bias to CBS news (e.g., Aday, Livingston, & Hebert, 2005). ABC and CNN ’3
framing of the Iraq issue also signiﬁcantly favored Kerry, as both these channels
devoted 10 more percentage points each to the pro-Kerry ﬂames than to the pro-
Bush ﬂames. By contrast, NBC’s ﬂaming seconds appeared to be the most
evenhanded, as the difference between pro-Kerry and pro-Bush ﬂames was only 2

percentage points, favoring Bush.

83

Table 13: Percentages of Iraq Issue Framing Seconds in News Channels.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Frames ABC CBS NBC CNN FOX MSNBC All
Ft terror 7% 5% 5% 3% 3% 9% 5%
Ft- 4% 4% 2% 3% 1% 2% 3%
ﬂeedom
Kerry 14% 12% 21% 20% 30% 29% 21%
inco.

Bush 2% 4% 10% 12% 4% 3% 7%

comp.

All pro- 27% 25% 38% 38% 38% 43% 36%

Bush (625) (589) (1,103) (1,903) (1,211) (694) (5,925)

Conﬂict 0% 2% 2% 0% 0% 0% 1%

Political 25% 25% 19% 13% 16% 5% 17%

ft.

Consensus 3% 1% 0% 0% 0% 0% 1%
All 28% 28% 21% 13% 16% 5% 19%

Neutral (664) (444) (630) (678) (498) (77) (2,991)
Policy fail 15% 17% 23% 26% 25% 26% 23%
Econ. 1% 3% 3% 3% 2% 2% 2%
cons.

. 15% 19% 5% 13% 14% 7% 12%
Bush 1nco.
Kerry 6% 3% 5% 6% 3% 5% 5%
com.
All pro- 37% 42% 36% 48% 44% 40% 42%
Kerry (834) (647) (1,061) (2,452) (1,384) (636) (7,063)
Other 8% 6% 6% 2% 2% 13% 5%
Total 100% 100% 100% 100% 100% 100% 100%

(2,350) (1,575) (2,957) (5,109) (3,149) (1,619) (16,759)

 

(Percentages are computed by columns. The sum of percentages may not exactly
equal 100 due to rounding. Numbers in parenthesis represent total ﬂaming
seconds. The columns give percentages of ﬂaming seconds by each channel and
the last column gives framing seconds for each ﬂame across all six channels).

84

 

In all, then, pro-Bush ﬂames ranged ﬂom a low of 25% (CBS) to a high of
43% (MSNBC), an 18 percentage point spread. Virtually all the news channels
gave the most attention to the pro-Bush Kerry incompetence ﬂame, which ranged
ﬂom 12% (CBS) to 30% (Fox). All other pro-Bush ﬂames were mostly under
10% of ﬂaming seconds (see Table 13), although CNN and NBC gave attention
also to the Bush competence ﬂame.

Less channel variance existed in use of pro-Kerry ﬂames, which ranged
ﬂom a low of 36 % (NBC) to a high of 48% (CNN), a 12 percentage point spread.
The policy failure/responsibility and Bush incompetence ﬂames were the most
ﬂequently used pro-Kerry ﬂames. Use of policy failure/responsibility ﬂame
ranged ﬂom 15% (ABC) to 26% (CNN). Use of the Bush incompetence ﬂame
ranged ﬂom 5% (NBC) to 19% (CBS). Other pro-Kerry ﬂames were used much
less ﬂequently, overall.

Interestingly, the networks were more likely to use the neutral ﬂames,
which ranged ﬂom (21%) (NBC) to 28% (ABC and CBS), compared to the cable
channels which ranged ﬂom 5% (MSNBC) to 16% (Fox) in use of the neutral
ﬂames. ABC and CBS were signiﬁcantly more likely to use the neutral ﬂames
compared to the cable channels, by the standards set for this study. MSNBC was
signiﬁcantly less likely than any other news program to use the neutral ﬂames.
One possible explanation for this is journalistic tradition in the networks
emphasizing impartiality and objectivity. In other words, focusing on the “horse
race” permits the news organization to “keep score” in a way that poses less

possibility of slipping into partisanship.

85

Network and Cable TV Diﬂerence in Framing

Hypothesis 1 posited that network and cable channels differed in the way
they ﬂamed the Iraq issue. It obtained partial support, but in only one of the three
assessments of ﬂaming seconds used. Both network and cable channels favored
Kerry. The data reveal that the cable channels were more favorable to the pro-
Kerry ﬂames (46%) than the pro-Bush frames (39%) by 7 percentage points. The
networks favored Kerry by 37% to Bush 31%, a ﬂaming difference of 6
percentage points (see Table 14). In terms of partisan balance, therefore, network
and cable channels hardly differed.

Data on the neutral ﬂames, however, reveal that networks gave
signiﬁcantly more time to the political ﬁght ﬂame (23%), used in only 13% of
cable channel ﬂames. Overall, the networks devoted 68% of their ﬂaming
seconds to pro-Bush or pro-Kerry partisan ﬂames and 25% to neutral ﬂames
compared to 85% of partisan frame time for the cable channels and 13% for
neutral ﬂame time.

This ﬁnding is in line with the argument that traditional broadcast
networks with thirty minute newscasts rely more on story types such as reporter
packages and anchor lead-ins to cover news (e.g., Wicks & Walker, 1993; also
see Table 3), which therefore increases the possibility that journalists stay on safer
ground by emphasizing horse race over issue coverage that may arouse partisan
ire. Cable channels’ newscasts are typically one hour long, which enables them to
report in longer news formats such as anchor in-house interviews (see Table 3)

thereby increasing the possibility of facilitating the ﬂames promoted by various

86

sources who may have overt or covert partisan agendas. Indeed, cable devoted
more ﬂaming time to the pro-Bush Kerry incompetence ﬂame than did the
networks (a 9 percentage point difference) and the pro-Kerry policy
failure/responsibility ﬂame than did the networks ( a 7 percentage point
difference) (see Table 14).

Table 14: Percentages of Framing Seconds in Network and Cable Channels.

 

 

 

 

 

 

 

Frames Networks Cable channels
All Pro-Bush 31% (2,117) 39% (3,808)
Fighting against terrorism 6% 4%
Fighting for ﬂeedom/dem. 3% 2%
Kerry incompetence 16% 25%
Bush competence 6% 8%
All Neutral 25% (1,738) 13% (1,253)
Conﬂict 1% 0%
Political ﬁght 23% 13%
Consensus 1% 0%
All Pro-Kerry 37% (2,592) 46% (4,472)
Policy failure/ resp. 19% 26%
Economic cons. 2% 2%
Bush incomp. 1 1% 13%
Kerry competence 5% 5%
Other 6% (435) 4% (344)
Total 100% 100%
46,882) (9,877)

 

 

 

 

(The sum of percentages may not exactly equal 100 due to rounding. Numbers in
parenthesis represent ﬂaming seconds).

Therefore, no substantial differences emerged when individual partisan
ﬂames were examined. The networks and cable stations virtually paralleled each

other in these emphases, but to different extents, in ﬂaming time.

87

Other ﬂames ﬂequently used by networks were the Kerry incompetence
(16%), and Bush incompetence (11%) ﬂames, while the cable channels also used
the Bush incompetence ﬂame (13%) (see Table 14).

Hypothesis 2 that predicted that the Fox News Channel (FNC) would be
more pro-Bush than other channels was not supported in either of the two
assessments of ﬂaming seconds used. The data in Table 13 reveal that FNC
favored Kerry by 6 percentage points, a frrrding contrary to expectations. In fact,
FNC lagged behind MSNBC in use of pro-Bush ﬂames, and simply equaled pro-
Bush ﬂames used in NBC and CNN. FNC was even more likely to focus Kerry
ﬂames than Bush ones, which made up 44% of its ﬂaming seconds. The contrary
ﬁndings for the FN C could account to the fact that this study was only examining
a speciﬁc newscast of the channel wherein it was possible that the channel was
focusing more on Kerry’s record rather than Bush. On the whole, if all other news
programs of FNC are considered then it might be possible to see more pro-Bush
coverage of the channel.

However, the ﬁndings of Hypothesis 2 are consistent with the Iraq war
study of Aday, Cluverius, & Livingston (2005), where they argued that FNC’s
coverage of the war was more in line with the three broadcast networks. Another
interesting observation with regard to FNC’s news coverage is that when it came
to devoting time to individual ﬂames, it gave its greatest attention to pro-Bush
Kerry incompetence ﬂame (30%) (see Table 13), signiﬁcantly more time than
given this ﬂame by ABC, CBS, and CNN. However, both CNN (20%) and

MSNBC (29%) also gave attention to this pro-Bush ﬂame.

88

Hypothesis 3 that posited that the CBS news used more pro-Kerry ﬂames
compared to other channels was partially supported in one of the two assessments
made of ﬂaming seconds, which is consistent with the ﬁndings of Katovsky &
Carlson (2003), and Lowry and Shidler (1995) that CBS News has more liberal
bias than other news channels. As noted earlier, CBS news devoted 42% ﬂaming
seconds to the pro-Kerry ﬂames versus 25% to the pro-Bush ﬂames, a difference
of 17 percentage points in favor of Kerry (see Table 13). No other channel had a
gap that large. CBS News with 42% pro-Kerry ﬂames was behind CNN (48%)
and FOX (44%) in devoting more ﬂaming seconds to those ﬂames. CBS News
gave less attention to pro-Bush ﬂames than did NBC, CNN, FOX, and MSNBC.

In temrs of individual pro-Bush and pro-Kerry partisan ﬂames, CBS was
not distinguished by signiﬁcant attention to particular ﬂames. CBS made the least
use of the pro-Bush Kerry incompetence ﬂame (12%), but not signiﬁcantly less
so than did three other channels. Correspondingly, CBS made the most use of the
pro-Kerry Bush incompetence ﬂame (19%), but not signiﬁcantly more so than did

three other channels.

Journalists and Their Frames

The second Research Question probed the kinds of ﬂames journalists
themselves used to cover the Iraq issue in statements they did not explicitly
attribute to any source. As indicated by the last column of Table 15, and
consistent with previous studies that journalists play up the political struggle
implications in news stories to make them appealing to the audience interested in

a ﬁght (e.g., de Vreese, Peter, & Semetko, 2001; Patterson & Donsbach, 1996;

89

Valkenburg et al., 1999), journalists overwhelmingly relied on the political ﬁght
frame (70%) to present the Iraq issue. For instance, ABC’s correspondent, Terry
Moran reported on October 7, 2004: “Peter, President Bush was on the defensive
on this [Iraq] issue today, and Senator Kerry is now using the president's handling
of Iraq as a primary reason to throw him out of ofﬁce.”

Journalists bound by professional and objective norms usually tend to
project such conﬂict aspects to make stories interesting and to attract the attention
of the audience. This may be the reason they ﬂamed the Iraq issue as a keenly
contested political ﬁght between the incumbent and the challenger. Moreover,
journalists norms for objectivity and impartiality mean that they mostly control
their own assertions and conﬁne their use of more partisan ﬂame to attributed

sources.

90

Table 15: Frequently Used Frames by Journalists.

 

 

 

 

 

 

 

Frames Network Cable All
All Pro-Bush 6% (136) 20% (339) 13% (475)
Ft. terrorism 0% 1% 1%
Ft. for ﬂeedom. 0% 0% 0%
Kerry incomp. 3% 16% 9%
Bush comp. 3% 3% 3%
All Neutral 73% (1,588) 69% (1,184) 71% (2,772)
Conﬂict 0% 0% 0%
Political ﬁght 71% 68% 70%
Consensus 2% 1% 1%
All Pro-Kerry 17% (363) 11% (193) 14% (556)
Policy fail/resp. 7% 2% 4%
Eco.cons. 2% 1% 2%
Bush incomp. 5% 7% 6%
Kerry compe. 3% 1% 2%
Other 4% (96) 0% (4) 3% (100)
Total 100% 100% 100%
(2,183) (1,720) (3,903)

 

 

 

 

 

(The sum of percentages may not exactly equal 100 due to rounding. Number in
parenthesis represent journalistic ﬂaming seconds).

Data in Table 15 also point at interesting differences in the kind of ﬂames
used by the network and cable channel journalists themselves to describe the Iraq
election issue (as opposed to ﬂames used in total coverage by all sources). The
journalists of networks used more of pro-Kerry ﬂames (17%) than pro-Bush
ﬂames (6%), which reﬂects a signiﬁcant difference of 11 percentage points
favoring Kerry in partisan balance. On the other hand, cable channel journalists
used more of pro-Bush ﬂames (20%) than pro-Kerry ﬂames (11%), an almost
signiﬁcant difference of 9 percentage points favoring Bush. In other words, the

partisan gap between network and cable journalists was 20 percentage points,

91

suggesting a signiﬁcant difference in the political orientation of cable and
network journalists. The network difference favoring Kerry, however, was
signiﬁcant by the standards used in this study, while the cable difference favoring
Bush was not. More speciﬁcally on particular partisan ﬂames, cable channel
journalists used the pro-Bush Kerry incompetence ﬂame (16%) more than did
network journalists (3%), a signiﬁcant 13 percentage point difference favoring
Bush. Network and cable channel journalists also differed in the use of other
individual ﬂames, but the differences were not signiﬁcant.

Findings showing the networks emphasizing more pro-Kerry frames is
consistent with the argument that they show a liberal bias. And cable channels’
favoring of pro-Bush ﬂames may result from the presence of FNC, which
according to researchers has a conservative bias. Therefore, this ﬁnding is
consistent with other studies (e.g., Aday, Livingston, & Hebert, 2005; D’Alessio
& Allen, 2000), which attribute liberal bias to networks and conservative bias to
FNC. With regard to use of neutral ﬂames, the networks’ emphasis for these
ﬂames (73%) was only 4 percentage points more than cable channels (69%) (see
Table 15).

Hypothesis 4 posited that journalists would primarily use political ﬁght
ﬂame to cover the Iraq issue. This hypothesis was supported, as journalists
overwhelmingly used the political ﬁght ﬂame (70%) while covering the Iraq issue
(see last column in Table 15), consistent with the studies of Patterson and

Donsbach (1996), and Valkenburg et al. (1999) and the discussion above.

92

Network journalists’ use of political ﬁght ﬂame (71%) was only 3 percentage
points more than that of cable reporters (68%) (see Table 15).

Research Question 3 explored whether race and gender of journalists
predicted differences in their ﬂaming of the Iraq issue. The difference between
white and non-white journalists was subtle when it came to ﬂaming of the Iraq
issue (see Table 16), which is consistent with similar ﬁndings of Zeldes and Fico
(2005). In terms of ﬂaming seconds, both white (70%) and non-white (66%)
journalists emphasized political ﬁght ﬂame, but the forrner’s ﬂaming seconds
percentage was slightly higher than the latter’s. Non-white journalists used
slightly more of pro-Kerry (18%) and pro-Bush (16%) ﬂames than did the white
journalists (see Table 16), but not signiﬁcantly so by the standards used in this
study.

In terms of overall balance, then, professional norms for impartiality
seemed to operate equally effectively for both white and non-white journalists.
The difference between white journalists’ use of the pro-Kerry ﬂames (14%) with
that of pro-Bush ﬂames (12%) was 2 percentage points, and this difference
between the pro-Kerry and the pro-Bush ﬂaming seconds is the same in the
coverage of non-white journalists (see Table 16). Non-white journalists
emphasized the pro-Kerry ﬂames with 18% ﬂaming seconds and 16% to the pro-
Bush ﬂaming seconds, but their use of these ﬂaming seconds was slightly higher
than the white journalists.

An example of a white journalist’s use of a balanced ﬂame is CBS’s Dan

Rather’s report on the Iraq issue: “Who should lead America, President Bush or

93

his Democratic challenger, Senator John Kerry? With security apparently the
number-one issue for voters nationally, the presidential candidates argued long
distance today over who would be a better commander in chief for the war on
terror, who has the better plan for the war in Iraq and the hunt for the still missing
Osama bin Laden. Thirteen days to go now, and we begin our coverage tonight
with CBS News correspondent Jim Axelrod,” (CBS, October 20, 2004). An
example of a non-white journalist’s use of political ﬁght ﬂame is CBS’s Byron
Pitts, who is African-American, Iraq issue report aired on September 16, 2004:
“While Senator Kerry sought to take political advantage of the president's
problems, President Bush ignored the intelligence report in his appearances today,

but not his opponent's own problems with Iraq.”

94

Table 16: Journalists’ Race and Framing Second.

 

 

 

 

 

 

 

 

 

 

Frames White journalists Non-white journalists
All pro-Bush 12% (416) 16% (59)
Ft. against terrorism 1% 1 %
Ft. for ﬂeedom/dem. 0% 0%
Kerry incompetence 9% 5%
Bush competence 2% 10%
All neutral 71% (2,530) 66% (242)
Conﬂict 0% 0%
Political ﬁght 70% 66%
Consensus 1% 0%
All pro-Kerry 14% (492) 18% (64)
Policy failure/resp. 3% 12%
Eco. cons. 2% 0%
Bush incompetence 6% 6%
Kerry competence 3% 0%
Other 3% (100) 0% (0)
Total 100% 100%
(3,538) (365)

 

(Numbers in parenthesis represent ﬂaming seconds).

With regard to individual ﬂames, white journalists tended to use Kerry
incompetence ﬂame (9%) slightly more than non-white journalists (5%). Non-
white journalists used policy failure/responsibility ﬂame (12%) more than did the
white journalists (3%). White and non-white journalists used same amount, 6%
each, of the Bush incompetence ﬂame seconds.

When compared to white journalists, non-white journalists did not use any
of ﬁghting for ﬂeedom and democracy, consensus, economic consequence, Kerry
competence, and other ﬂames (see Table 16). In line with Gandy et al.’s (1997)
argument, the data here indicate that non-white journalists were willing to use

ﬂames different than did their white counterparts, thereby providing a different

95

angle or viewpoint in the Iraq issue. Non-white journalists, however, also did not
use at all some of the ﬂames used by white journalists. Again, however, these
differences were not signiﬁcant by the standards set for this research.

Hypothesis 5 predicting that women journalists would use more pro-Kerry
ﬂames was supported (see Table 17), consistent with a study of Gidengil and
Everitt (2003). When it came to the partisan balance in their ﬂaming seconds,
women journalists used more of pro-Kerry ﬂames (27%) than the pro-Bush
ﬂames (17%), a signiﬁcant difference favoring Kerry. For instance, NBC’s Kelly
O’Donnell while covering Kerry’s campaign on September 8, 2004 said: “Tonight
a more aggressive John Kerry is trying to reﬂame the discussion on the president's
signature issue. Kerry's new buzzword, "choices," urging voters to see the costs of
the war in Iraq differently.” The data in Table 18 also reveal that male journalists
emphasized equal amounts, 11% each, of the pro-Bush and the pro-Kerry ﬂaming
seconds, compared to the 10 percentage point difference between the pro-Kerry
(27%) and the pro-Bush (17%) ﬂaming seconds for the women journalists. For
instance, while covering the Iraq issue on September 8, 2004, CBS’s Jim Axelrod
balanced the opposing partisan ﬂames: “Eight weeks to Election Day, Iraq
remained topic one in the race for the White House. . .While the president
provided his latest rationale for the war, his challenger argued it's actually created
more terrorists.” Therefore, consistent with earlier studies (Gidengil & Everitt,
2003; Sutcliffe et al., 2005), use of more pro-Kerry ﬂames by women journalists
indicate that they were critical of the Bush administration’s handling of the Iraq

issue.

96

The data in Table 17 also answers a part of the third Research Question
seeking to ﬁnd the inﬂuence of gender difference on the ﬂaming of the Iraq issue.
Keeping with journalistic norms, women reporters did primarily use neutral
ﬂames (57%) in the coverage of the Iraq issue. However, they used those ﬂames
signiﬁcantly less than did their male colleagues.

One major difference between male and female journalists in the use of
partisan ﬂames was that the latter used the Bush incompetence ﬂame (15%)
signiﬁcantly more than did men (4%) (see Table 17). Apart ﬂom this major
difference in use of the Bush incompetence ﬂame, both male and female
journalists were similar in using remaining ﬂames, which is consistent with the
study of Sutcliffe et al. (2005), that the gender difference in the news coverage is
subtle. Another difference was that female journalists made no use at all of two
pro-Bush ﬂames. In line with the arguments of Sutcliffe et al. (2005), and
Gidengil and Everitt (2003), the data in Table 17 reveal that women reporters
were more critical of the administration and expressed concern on the Iraq issue
by using more of the Bush incompetence frame than their male counterparts.

It is unclear why there should be such a “gender gap” in Iraq war ﬂaming,
especially in the light of the nearly equal attention given to the partisans by white
and non-white journalists. This ﬁnding may reﬂect a gender-based partisanship
that is present in newsrooms and that is becoming more visible as women enter
more powerful management positions. Possibly, too, it may reﬂect more
autonomy and less scrutiny given female journalists. Finally, such a ﬁnding could

indicate that female journalists may be less experienced and have not been ﬁrlly

97

socialized into occupational objectivity norms. Unfortunately, this study can only

speculate on these possible explanations rather than establish a deﬁnitive one.

Table 17: Journalists’ Gender and Framing Seconds.

 

 

 

 

 

 

 

Frames Male journalists Female journalists
All pro-Bush 11% (342) 17% (133)
Ft. against terrorism 1% 0%
Ft. for ﬂeedom/dem. 0% 0%
Kerry incompetence 8% 1 1%
Bush competence 2% 6%
All neutral 75% (2,317) 57%(455)
Conﬂict 0% 0%
Political ﬁght 74% 55%
Consensus 1% 2%
All pro-Kerry 11% (339) 27% (217)
Policy failure/resp. 4% 5%
Eco. cons. 1% 5%
Bush incompetence 4% 15%
Kerry competence 2% 2%
Other 3% (100) 0% (0)
Total 100% 100%
(3,098) (805)

 

 

 

 

(The sum of percentages in columns may not exactly equal 100 due to rounding.
Numbers in parenthesis represent total ﬂaming seconds).

Partisans and Framing

Hypothesis 6 posited that stories with a greater Republican partisan
balance would use more pro-Bush ﬂames. Not surprisingly, this hypothesis was
supported, consistent with the studies of Fico and Cote (2002), Jacoby (2000), and
Parmelee (2002). The data in Table 18 reveal that stories favoring Bush (in other

words, stories having a Republican partisan tilt), got a high percentage (63%) of

98

pro-Bush framing seconds, while the remaining pro-Kerry, neutral or other ﬂames
got less then 20% each. Therefore, consistent with earlier studies (Fico & Cote,
2002; Jacoby, 2000; Parmelee, 2002), the data in Table 18 also indicate that
during election campaign stories favoring Republican sources with more
prominent attention were dominated by the ﬂames projected by them.

Hypothesis 7, correspondingly predicting that stories having greater
Democratic balance would use more pro-Kerry frames was also supported (see
Table 18), as expected. In the stories favoring Kerry, 69% pro-Kerry ﬂames were
found, while the remaining ﬂames again were less than 20%. Again, this ﬁnding
was consistent with the studies of Fico and Cote (2002), Jacoby (2000), and
Parmelee (2002) that stories dominated by a particular partisan group, in this case
Democrats, will have more of the ﬂames projected by that partisan group.

Table 18 also, of course, illustrates what happens to stories that are not
exacting in their standards of “fairness and balance.” Certainly, including even
one source opposed to the others favoring a candidate will make a story more fair.
Absent nearly an equal number of sources on both sides, however, the story will
not be more even. Moreover, departures ﬂom even treatment in prominence given

candidate ﬂames predicts more of the same in the balance of the story.

99

Table 18: Partisan Story Domination" and Percentage of Frames.

 

 

 

 

 

 

Frames Stories favoring Bush Stories favoring Kerry
Pro-Bush 63% 15%
(3,771) (1,086)
Neutral 17% 14%
(993) (987)
Pro-Kerry 17% 69%
(1,034) (4,969)
Other 4% 2%
(221) (157)
100% 100%
Total (6,019) (7,199)

 

 

 

 

 

*Stories favoring Bush or Kerry sources could also have sources supporting the
opponent included. (The sum of percentages may not exactly equal 100 due to
rounding. Number in parenthesis represent Republican and Democratic partisan
ﬂaming seconds).

(Pro-Bush ﬂames: ﬁghting against terrorism, ﬁghting for freedom and
democracy, Kerry incompetence, and Bush competence; Pro-Kerry ﬂames: Policy
failure/responsibility, Economic consequence, Bush incompetence, Kerry
competence; Neutral: conﬂict, political ﬁght, and consensus; Other).

The related Research Question 4 sought to ﬁnd out whether Republican or
Democratic sourced ﬂames were more likely to be found in the coverage. The
data in Table 19 reveal that Democratic sourced pro-Kerry ﬂames (94%) were
slightly more dominant than the Republican-sourced pro-Bush (92%) ﬂames. This
ﬁnding suggests that the Republican and Democratic sources keenly contested
with each other to project their ﬂames on the Iraq issue in the news media. There
were few “mavericks” in either party who criticized their party’s candidate or who
failed to target the opponent. With regard to individual ﬂames, the Republican-
sourced Kerry incompetence ﬂame (53%) and the Democratic-sourced policy

failure/responsibility ﬂame (51%) were most ﬂequent ﬂames in the coverage of

the Iraq issue in the 2004 presidential race (see Table 19).

100

Table 19: Percentages of Republican and Democratic Sourced Frames.

 

 

 

 

 

 

 

Frames Republican sources Democratic sources
All pro-Bush 92% (4,994) 2% (136)
Ft. against terrorism 13% 1%
Ft. for ﬂeedom/dem. 8% 0%
Kerry incompetence 53% 1%
Bush competence 18% 0%
All neutral 2% (76) 2% (106)
Conﬂict 1% (41) 1%
Political ﬁght 0%(9) 0%
Consensus 1% (26) 1%
All pro-Kerry 2% (1 13) 94% (5,913)
Policy failure/resp. 1% 51%
Eco. cons. 0% 5%
Bush incompetence 1% 27%
Kerry competence 0% 11%
Other 5% (264) 2% (132)
Total 100% 100%
(5,447) (6,284)

 

 

 

 

(Frames with preﬁx All represent a sum of sub-ﬂames mentioned under them in
each row. The sum of percentages in columns may not exactly equal 100 due to
rounding. Number in parenthesis represent Republican and Democratic sourced

framing seconds).

This reﬂects an astonishingly effective ability of both parties to stay “on
message” during the campaign. For both campaigns, the most ﬂequent ﬂaming
was negative. However, the three other Bush ﬂames, making up 39% of the
Republican—sourced ﬂames, can be interpreted as positive, emphasizing
protection against terrorism, the ﬁght for ﬂeedom and president’s competence.
For the Kerry campaign, another 27% of the ﬂaming second also went negative,

emphasizing the president’s alleged incompetence.

101

Overall, then, while both campaigns stressed negative aspects, the Bush
campaign was more positive. Even when it came to the positive ﬂaming of the
competence of their own candidate, Republicans were more likely to do that (18%
of ﬂaming seconds) than were Democrats (11% of ﬂaming seconds) (see Table
19). Of course, it is also possible that journalists distort political reality by
selectively using ﬂames out of proportion to the candidates’ use of them. This
study, however, cannot address that possibility.

Non-partisans and Framing

Research Question 5 asked what kind of ﬂames were most often used by
experts and non—partisan sources. As indicated by Table 20, non-partisan sources
hardly ever used the neutral ﬂames. Although they were not nearly as partisan in
their ﬂaming as declared Republican or Democratic sources, they could not be
characterized as neutral.

Overall, non-partisans most ﬂequently used pro-Kerry ﬂames (43%)
followed by pro-Bush ﬂames (28%), a signiﬁcant 15 percentage point difference
between those ﬂames. With regard to individual ﬂames, the policy
failure/responsibility ﬂame (3 5%) was most ﬂequently used, followed by Kerry
incompetence ﬂame (24%). For example, the pro-Kerry policy
failure/responsibility ﬂame was used by an expert on the Iraq issue, Bathsheba
Crocker: “...Issues like security and governance and basic services. We found a
fairly depressing picture across these range of issues, and we found that things are

not really headed in the right direction yet [in Iraq],” (NBC, September 8, 2004)

102

As indicated by these results, non—partisan sources were primarily critical
of the Bush administration’s handling of the Iraq issue. So even though non-
partisans did not clearly side either with Republicans or Democrats, the ﬂames
they tended to use more often supported Kerry’s agenda. This may reﬂect
“covert” partisanship on the part of these sources, or a failure of journalists to
fully identify the partisan ties of these sources. Of course, this ﬂaming pattern
may also reﬂect the expert assessments of genuinely non-partisan sources.
Unfortunately, data ﬂom this research cannot determine which explanation ﬁts

best.

Table 20: Frames Used by Expert/Non-partisan Sources.

 

 

 

 

 

 

 

Frames Percentages
of ﬂames
All Pro-Bush 28% (320)
Fighting against terrorism 0%
Fighting for ﬂeedom/dem. 0%
Kerry incompetence 24%
Bush competence 4%
All Neutral 4% (40)
Conﬂict 0%
Political ﬁght 4%
Consensus 0%
All Pro-Kerry 43% (482)
Policy failure/ resp. 35%
Economic cons. 0%
Bush incompetence 5%
Kerry competence 3%
Other 25% (83)
Total 100%
(1,125)

 

 

 

(The sum of percentages in columns may not exactly equal 100 due to rounding.
Number in parenthesis represent total non-partisan ﬂaming seconds).

103

Summary of Research Questions and Hypothesis Test Results

For the purpose of this study, 445 news stories were analyzed from three
networks and three cable channels. Stories with dominant ﬂames were
signiﬁcantly more likely to favor Kerry than Bush. But when all story ﬂaming
seconds were considered, that gap closed to a 6 percentage point difference.

A close look at the ﬂaming seconds of individuals channels also revealed
some signiﬁcant differences among them. CBS devoted more ﬂaming seconds to
pro-Kerry ﬂames (42%) than pro-Bush ﬂames (25%), leading all the channels.
ABC and CNN were also signiﬁcantly more likely to favor Kerry over Bush,
according to the standards used in this study. No channel was signiﬁcantly more
likely to favor Bush in its Iraq war ﬂaming.

RQl asked about the relative ﬂequency of ﬂames in coverage. Overall,
few ﬂames exceeded 10 percent of stories on seconds. Focus was given largely to
negative ﬂames such as the pro-Bush Kerry incompetence ﬂame, and pro-Kerry
policy failure and Bush incompetence ﬂames.

Hypothesis 1 was partially supported, ﬁnding that cable channels made
more use of partisan than neutral ﬂames in their coverage compared to the
networks (see Table 15).

Hypothesis 2 that predicted that the Fox News Channel (FNC) would use
more pro-Bush ﬂames than other channels was not supported, as FNC’s ﬂaming
of the Iraq issue actually favored the pro-Kerry ﬂames by 6 percentage points and
did not differ signiﬁcantly ﬂom other channels in their use of particular partisan

ﬂames.

104

Hypothesis 3 that posited that the CBS News used more pro-Kerry ﬂames
compared to other channels was partially supported. CBS News devoted 17
percentage points more to pro-Kerry ﬂames than pro-Bush ﬂames. Among all six
channels, CBS news devoted less time to ﬂames supporting Bush on the Iraq war,
but not signiﬁcantly so in all comparisons.

RQ 2 probed the kinds of ﬂames used by the jornnalists in their own
words to ﬂame the Iraq issue. Journalists overwhelmingly used the political ﬁght
ﬂame (70%). Their use of pro-Bush (13%) and pro-Kerry (14%) ﬂames was
almost equal. Network journalists used more pro-Kerry ﬂames (17%) than pro-
Bush ﬂames (6%), whereas cable channel reporters used more pro-Bush ﬂames
(20%) than the pro-Kerry ﬂames (11%). Cable and network journalists therefore
differed in their political orientations, at least by the measures and standards used
in this research.

Hypothesis 4, related to the second Research Question that posited that
journalists would primarily use political ﬁght ﬂame, was supported as journalists
overwhelmingly used that ﬂame (70%).

RQ3 probed the inﬂuence of race and gender of journalists on the ﬂaming
of the Iraq issue. Race was not signiﬁcant, but gender was. Non-white journalists
used slightly more of pro-Kerry (18%) and pro-Bush (16%) ﬂames than the white
journalists, but the race of journalists was associated with few and negligible
differences. Female journalists, however, devoted more time to pro-Kerry ﬂames
(27%) than pro-Bush ﬂames (17%), whereas male reporters devoted equal time

(11% each) to pro-Bush and pro-Kerry ﬂames. In other words, women journalists

105

were more partisan and Democratic in their political orientation, according to
these data.

The related Hypothesis 5 that female journalists would use more pro-
Kerry frames was therefore supported.

Hypothesis 6 and 7 that posited that Republicans and Democrat partisan
sourced stories would be dominated by pro-Bush and pro-Kerry ﬂames were also
supported.

The related Research Question 4 sought to ﬁnd out whether Republican or
Democratic sourced ﬂames were likely to be found in the coverage. Democratic
sourced pro-Kerry ﬂames (94%) were only slightly more dominant than the
Republican-sourced pro-Bush (92%) ﬂames.

Finally, Research Question 5 probed what kind of ﬂames were most often
used by experts and non-partisan sources. Non-partisans ﬂequently used more
pro-Kerry ﬂames than pro-Bush ﬂames.

Assessment of Relative Strength of Inﬂuences

Given these ﬁndings, an assessment can also be made about the relative
power of various sources to set story ﬂames. Certainly this is related to the
numbers of sources of ﬂames and the time given them in stories.

In sum, then, partisans were clearly the most important ﬂame setters found
in this study. They were the most numerous ﬂame-setting sources. Not
surprisingly, the ﬂames they set favored the supported candidate, but with some
important differences. Bush ﬂames were more positive and Kerry ﬂames more

negative.

106

Next to partisans, channel type, a proxy for news management practices at
network and cable channels, was the next most powerful inﬂuence. Speciﬁcally,
the traditional networks were most likely to hold to their more neutral and
impartial role, emphasizing political horse race coverage ﬂames more than did
their cable colleagues.

The gender of journalists had a much smaller but striking difference on
frame setting. Certainly the size of their ﬂaming inﬂuence was severely limited
by the fewness of their numbers, and, therefore, the relatively low ﬂequency of
their ﬂaming assertions. Nonetheless, women journalists were more likely to bias
stories with ﬂames more supportive of Kerry than Bush.

Non-partisan sources were least inﬂuential on ﬂame setting, but only
because so few of them were used relative to partisans and journalists. Like
women journalists, however, their ﬂame setting was more likely to facilitate the
Kerry campaign than the Bush one.

The regression analysis mostly gave emphasis to these ﬁndings (see Table
21), but are speciﬁc to imbalance in attention to pro-Bush or pro-Kerry ﬂames.
The regression analysis therefore applies to a much narrower question about
framing in election coverage.

The most important factor producing ﬂaming imbalance in stories was the
use of partisan sources (compared to ﬂaming ﬂom journalists and non-partisan
sources). As indicated by Table 21, the unstandardazied coefﬁcient for partisan
vs. all other ﬂame sources was 20.1. In other words, use of partisan sources

increased story imbalance by an average of 20.1 seconds. While not a surprising

107

ﬁnding, this also suggests that stories including other sources depresses such
imbalance.

Next in inﬂuence, stories produced by network news programs had more
overall imbalance in partisan ﬂaming than did cable channels, even though study
data show that networks used more neutral ﬂames than cable channels. The data
indicate that if the story is produced by networks then the story imbalance
increases on an average by 5.8 seconds. This may be due, in part, to the greater
emphasis network news placed on covering the presidential challenger compared
to cable channels.

Reporter gender (.37) and race (1.5) produced much smaller inﬂuences on
imbalance. Stories reported by women journalists were less imbalanced in their
ﬂaming, even though study data show that the direct ﬂaming statements of
women reporters tended to be more supportive of the Kerry campaign.

Overall, these four variables explained about 20% of the variance in
ﬂaming imbalance, although source type and channel type accounted for much of
that imbalance. Obviously a great many other factors not included in this study

are also inﬂuential on such imbalance.

108

Table 21: Inﬂuence of Factors on Partisan Frame Imbalance. Coefﬁcients are
Betas. Numbers in parenthesis are unstandardazied coefﬁcients.

 

 

 

Level and types Iraq issue partisan
of inﬂuence ﬂame imbalance

Network vs. cable -.13 (-5.8)
news
White vs. non-white .02 (1.5)
journalists
Male vs. female -.06 (-.37)
journalists
Partisan vs. all other .40 (20.1)
ﬂame sources

 

 

 

R2 = .21; Equation Signiﬁcance = .000

Scaling of independent variables: networks = 1, cable channels = 0; white
journalists = 1, non-white journalists = 0; male journalists = 1, female journalists
= 0; partisan sources = 1, all other sources = 0.

109

CHAPTER 6
Discussion

The ﬁndings of this study added speciﬁc knowledge to the framing
literature by showing theoretical and empirical evidence of power of partisan
sources to set ﬂames for an issue like the Iraq conﬂict during presidential race.
The data of the study also showed the difference in ﬂaming of stories by network
and cable channels. Based on the past research and ﬁndings of this study, it is
suggested that issue-speciﬁc ﬂames should be anchored in generic ﬂames such
as conﬂict, responsibility, and human interest, etc. Anchoring issue speciﬁc
ﬂames into generic ﬂames would facilitate cross-issue generalizability, thereby
making ﬂaming studies more robust in their prediction, explanation, and control
of a phenomenon in mass communication.

As noted earlier, this study examined the media ﬂaming of the Iraq issue
because it was one of the top issues during the 2004 presidential campaign. Even
almost two years after the election, the Iraq issue continues to be a thorn in the
American polity. With the kind of uncertainty looming over the Iraq policy, it
may not be a surprise to see this issue again on the ballot in the 2008 presidential
election. How America as a nation was divided over this issue during the 2004
presidential race was evident ﬂom CNN ’8 interview of Cindy Sheehan and Bill
Buesing (CNN, 2004). Sheehan and Buesing had lost their sons in the Iraq war,
but the latter supported the war while the former opposed it. The Iraq issue, with a

nation sharply divided on it, ﬂom a mass communication perspective calls for

110

examining the news media’s role in covering and ﬂaming of this issue as it had
implications for public opinion beyond the presidential race.

From a journalism point of view, scholars (e.g., Aday, Livingston, &
Hebert, 2005; Norris, Kern, & Just, 2003; Ross & Bantimaroudis, 2006) have
argued that aﬂer the September 11 attacks American journalism has undergone
noticeable change, claiming for instance, that the news media professionals are
raising above objective norms to freely express patriotic fervor, and rally around
government sources thereby projecting more of their ﬂames on Middle East
violence and terrorism. For instance, Aday, Livingston, & Hebert (2005) pointed
out: “Since the terrorist attacks of September 11, journalistic objectivity has been
a matter of ﬂequent controversy. . .Everything ﬂom whether anchors should wear
patriotic lapel pins to how much a network should show civilian casualties has
been at issue. Not just administration ofﬁcials but journalists themselves have
berated news organizations deemed not patriotic enough, with some suggesting
that there is no place for detachment in wartime,” (p. 6). Moreover, according to
Entman (2004), in this post-Cold War era the September 11 attacks provide a
larger context to understand the role of the media. This also gives a reason to
examine ﬂaming of the Iraq issue in the ﬁrst presidential campaign after the
September 11 attacks, and for assessing the theoretical and practical implications
of such coverage. Speciﬁcally, what can be learned to better predict and explain a
communication process? And how might journalism better ﬁrlﬁll its normative

role?

11]

Implications for Theogy
Reassessing The Model of This Study

From a theoretical perspective, this study examined the type and
prevalence of media ﬂames in television news with regard to the Iraq issue. This
study analyzed how various communication variables inﬂuenced the ﬂaming of
the Iraq issue in news media in the context of 2004 presidential campaign. The
theoretical springboard for this study was provided by an argument made in
Scheufele’s (1999) widely cited work on ﬂaming effects. Citing Gans (1979),
Shoemaker and Reese (1996), and Tuchman (1978), Scheufele (1999) argued that
studies have been done in the past to examine the inﬂuence of extrinsic and
intrinsic media factors on production and selection of news, while “evidence has
yet to be systematically collected about how various factors impact the structural
qualities of news in terms of framing,” (p.109). Therefore, this study responded
to that call by systematically examining how selected factors inﬂuenced the
structural qualities of news ﬂaming. It adds to a growing literature exploring the
inﬂuence of various communication variables on media ﬂames. Identifying and
studying media ﬂames is important because they have implications for public
opinion, especially during elections when candidates keenly contest on key issues.

Medium type, deﬁned as network versus cable news outlets, was
considered to assess whether they differed in their inﬂuence on the Iraq ﬂames.
The theoretical assumption here was that medium type would inﬂuence ﬂaming
as cable channel newscasts are longer than networks’, allowing them to cover and

ﬂame news differently. The study data suggested that the networks used more of

112

the political ﬁght ﬂame when compared to the cable channels, reﬂecting that the
former emphasized more the traditional “horse race” coverage.

This study also took into account the race and gender of journalists to
examine their inﬂuence on ﬂaming. The assumption here was that race and
gender of a journalist would mediate their ﬂames on the Iraq issue. According to
the research ﬁndings of this study, gender did inﬂuence ﬂaming to some extent,
but race did not.

Finally, sources, both partisans and non-partisans including journalists
themselves were examined to see what kinds of ﬂames they were sponsoring on
the Iraq issue. The study data revealed that ﬂames sponsored by partisans
overwhelmingly dominated the Iraq issue.

This study organized these inﬂuences based on Shoemaker and Reese’s
(1996) hierarchical model in which media factors such as nature of the medium
(cable vs. network channels) inﬂuenced the race and gender of journalists
working for news organizations which in turn inﬂuenced the use of partisan and
non-partisan sources on news media ﬂaming of the Iraq issue. This study
contributes to knowledge in this area by empirically testing a model that takes
account of medium, journalist, and source inﬂuence on ﬂaming in a policy
conﬂict context.

As noted earlier, this model was robust in showing strong inﬂuence of
partisan sources on news media ﬂaming of the Iraq issue. It was moderately

successful in showing the inﬂuence of medium and journalist gender (but not

113

race) on the Iraq ﬂames. It replicated past work showing the inﬂuence of partisans
not only broadly on the news, but speciﬁcally on public policy issue ﬂaming.

On the whole, this study found the hierarchical model of Shoemaker and
Reese (1996) was helpful in exploring the inﬂuence of nature of medium, race
and gender of journalists, and partisan and non-partisan sources (including
journalists) on news media ﬂaming of the Iraq issue.

This model, however, could be strengthened by increasing the generality
of the ﬂaming dependent variable. The model can also be improved by using
variables with additional explanatory and predictive power. These modiﬁcations
are discussed below. But before discussing the revised model, it would be helpﬁrl
to ﬁrst try to resolve the theoretical debate over generic versus issue-speciﬁc
ﬂames in the light of this study’s approach.

Using Generic and Issue Speciﬁc Frames

While one should be cautious to generalize the results ﬂom a single issue
study to a general theory of public policy debate formation, the issue—speciﬁc
research ﬁndings of this study yielded rich details. Study data suggest that
journalists primarily served as conduits for elite opinions and ﬂames. Most of
their Iraq issue coverage was sourced by partisan elite sources, while non-
partisans such as experts and ordinary citizens were hardly used as sources. This
is the reason partisan ﬂames sponsored by pro-Bush and pro-Kerry sources
dominated the Iraq issue stories. Moreover, the issue speciﬁc context for this
study points to reasons for this pattern. An explanation justifying ﬁnding more

partisan ﬂames in the Iraq issue coverage is that during special events like

114

presidential race, news media tend to focus more on the presidential candidates
and their issue positions. Further, as the Iraq issue was being covered in the
election context, the journalists themselves ﬂamed this issue as a political ﬁght
between the rival presidential candidates.

However, this political ﬁght ﬂame itself was part of a larger generic,
strategic ﬂame used by the journalists to cover many issues applied to the Iraq
issue, it allowed them to report objectively by emphasizing the “horse race.” In
other words, the political ﬁght ﬂame allowed journalists to tell who was winning
and losing without sounding or appearing biased.

In this study, the issue speciﬁc ﬂaming of the Iraq policy, examined in the
presidential campaign context, helps contribute to the issue ﬂaming approach. In
this research, it was appropriate to use the issue speciﬁc ﬂarrring as it facilitated
the investigation of the Iraq issue in a greater detail, which otherwise would not
have been possible with the mere use of generic ﬂames. This study attempted to
capture speciﬁc aspects of selection, organization, and elaboration that were
present in the coverage of the Iraq issue.

But it is also possible to see how this study ﬁts into a generic ﬂame
context. This study used Entman’s (1993) deﬁnition that “to ﬂame is to select
some aspects of a perceived reality and make them salient in a communicating
text, in such a way as to promote a particular problem deﬁnition, causal
interpretation, moral evaluation and/or treatment recommendation,” (p. 53). For
instance, the fighting against terrorism and the ﬁghting for freedom and

democracy ﬂames used in this study are examples of how the Bush administration

115

was defming the Iraq issue problems and solutions. The Kerry camp’s ﬂames of
policy failure/responsibility, and economic consequence are examples of causal
interpretation and attributing responsibility to the Bush administration for creating
a mess in Iraq. While the Bush incompetence and Kerry incompetence frames tie
into the moral evaluation category, the Bush competence and Kerry competence
frames are examples of solution recommendations for the Iraq issue. In general,
therefore, this study has found these more generic categories of thought to be
appropriate and important in the context of an intense political and public policy
contest. Indeed, the ﬂames found empirically in this research ﬁt well into
Entman’s categories.

This raises the possibility that issue speciﬁc ﬂames could and should be
integrated into generic ﬂames, thereby increasing overall the generality of
ﬂaming research. For instance, de Vreese et al., (2001) argue that issue-speciﬁc
ﬂames are applicable to speciﬁc topics or news events, whereas generic ﬂames
are broadly applicable to a range of different news topics, some even over time,
and potentially, in different cultural contexts. An issue-speciﬁc news ﬂame allows
for investigation of the ﬂaming of particular events in greater detail. It may
capture speciﬁc aspects of selection, organization, and elaboration that are present
in news coverage and pertain speciﬁcally to a well-defmed issue. However, the
greater detail and issue-sensitivity renders issue speciﬁc analyses difﬁcult to
generalize, capture, and use as a base for general hypothesis and theory building.
On the other hand, generic ﬂames offer less possibility for examining the ﬂaming

of an event in greater detail, but they allow comparisons between ﬂames, topics,

116

and ﬂaming practices within and between different media outlets (de Vreese et
al., 2001).

This study acknowledges that the speciﬁc Iraq issue ﬂames identiﬁed in
this research are irrelevant to other issues. But the issue categories of this study,
based on Entman’s (1993) ﬂaming categories, can be applied to other issues to
examine how they were ﬂamed. For instance, the empirical evidence of this study
on the Iraq issue cannot be directly applied to an issue like abortion. However,
Entman’s (1993) ﬂaming categories, can be applied to both the abortion issue as
well as others to examine the relative way in which issues are being categorized
as problems, causally interpreted, morally evaluated, and solved. The two
different issues—and others——can then be compared for variations in how these
generic ﬂames were used. Also, these generic ﬂames enrich ﬁrture studies of
similar issues such as ﬂaming of the war on terrorism, ongoing debate on the Iraq
issue or comparing media ﬂaming of the ﬁrst and the second Gulf Wars.

Further, to increase cross-issue generalizability of the issue-speciﬁc
frames even more, speciﬁc studies can be tied to broader generic ﬂames that are
replicable across studies, such as conﬂict, human interest, responsibility,
economic consequence, and morality, which have been found to a great extent in
the US. news media (Semetko & Valkenburg, 2000). For instance, the Iraq issue
can be examined under a broader generic ﬂame of conﬂict and then under a
“middle level” set of generic ﬂames such as those supplied by Entman (1993).

Similarly, abortion can be studied under a broader ﬂame like morality. Once

117

issues are linked to broader generic ﬂames, then, by using Entman’s (1993)
ﬂaming categories, they can be examined across studies.

Thus, to develop a common theoretical ﬂame, future studies should anchor
speciﬁc issues in generic ﬂames deliberately chosen to permit comparison across
very different issues. And more issues should be examined, because, as argued by
Bennett (1990), “only by exploring different cases can we see the general
tendencies of press state relations and thus begin to proﬁle the operations of US.
democracy across different issue areas,” (p. 20).

Speciﬁcally, then, the generic ﬂames should be held constant across
studies in a research program that varied in attention to speciﬁc issues and ﬂames
particular to them. Doing this gains generality at the generic level and detail at the
issue-speciﬁc level. The dependent variable in the revised model below (see
Figure 2) attempts to illustrate these two levels, emphasizing a limited number of
generic ﬂames useful across a wide range of studies dealing with the widest
possible range of speciﬁc issues or events.

The revised model in this study (see Figure 2) pr0poses that future ﬂaming
research should use the following ﬁve generic ﬂames—conﬂict, human interest,
responsibility, economic consequence, and morality. The reason for using these
ﬁve ﬂames is that past research has emphasized existence of these ﬁve ﬂames in
the news. The wide use of these ﬁve generic ﬂames was also suggested by widely
cited work of Semetko and Valkenburg (2000). Moreover, for a scientiﬁc study,
measures that can be generalized and replicated across issues and events are

required to predict, explain and control phenomena. And these ﬁve ﬂames have

118

been found to have cross-issue generalizability, as they have been applied to a
variety of issues such as the Iraq war (Dimitrova et al., 2005; Luther & Miller,
2005), European politics (Semetko & Valkenburg, 2000), international
comparative studies (de Vreese, Peter, & Semetko, 2001), political campaigns

(Druckrnan, 2005), and even science related ﬁelds (Dahinden, 2002).

Revised Model of Inﬂuences on Framing

 

Media Type
Print vs. Broadcast
Elite vs. Non-elite
Traditional vs. Blogs

 

 

 

 

Generic Frames
Conﬂict
Human Interest
Responsibility
News Gatherer Econ. Cons.
Resources Morality

 

 

 

 

 

 

 

 

Source Use . memﬁ

 

 

 

 

 

Issue E.g.:
News Gatherer Iraq

Diversity Economy/J ob
Abortion
Celebrations
Sex Roles
Fashion

Etc.

 

 

 

 

 

 

 

 

Gender Ratio

 

 

 

Figure 2: Revised Model of Inﬂuences on Framing: Revised model explicating
inﬂuence of communication variables on generic and issue speciﬁc ﬂames
mediated by source use.

119

For instance, if framing of an issue like economy/job is being examined,
then the economic consequence ﬂame might be “set” as generic ﬂame for this
issue. The researchers, then, could take a closer look at the messages on the
economy/job to determine the issue speciﬁc ﬂames. Such an approach permits an
analysis of many issues (in this case, relating to economic topics) in many studies
as they relate to direct generic ﬂames, while permitting detailed study of issue
speciﬁc ﬂames.

Given that the goal of science is to predict, explain and control, this can be
done at the level of generic ﬂaming. For instance, if Entman’s (1993) ﬂaming
categories are considered for generic ﬂames, then researchers can design a study
which facilitates cross-issue generalizability and prediction about the kinds of
issues likely to have particular distributions across the Entman categories. An
example for such a study could be analyzing how issues of health care and gun
control, under generic ﬂames, were deﬁned as problems, how responsibility was
assigned for these problems, whether these issues were tied to a morality debate,
and what solutions were prescribed for these issues. Thus, although these issues
are not related, they can still be examined under generic ﬂames that facilitate
generalizability. And once these generic ﬂames are established, then each of these
issues can be examined in detail for ﬂames speciﬁc to them.

Predictors of Media Framing

Clearly too, the independent variables in the model can be modiﬁed in the
light of this study’s ﬁndings. In general, it is suggested that media type as an

explanatory variable should be expanded in its values, while the journalistic

120

variables used in this study at the individual level should be redeﬁned as
newsroom factors. Moreover, given that ﬂaming involves a power play not only
among political interests, but also between those interests and journalism
institutions, a measure of news organization power should be added to the model.
These modiﬁcations are discussed below.

As per the revised model (see Figure 2), new independent variables such
as media type—elite versus non-elite, traditional versus blogs—could be added to
assess their inﬂuence on the media ﬂames. While these variables can have direct
inﬂuence on the media ﬂames, their inﬂuence can also be mediated through an
intervening variable, source use. For instance, by adding elite media like the New
York Times and the Washington Post and non-elite media such as local
newspapers, this model assumes that they would use different sources who would
ﬂame issues differently. The elite media might reﬂect the ﬂaming of the issues at
a broader, national level, where as local newspapers would reﬂect issue ﬂaming
in local communities. In some instances it is also a possible that both national and
local media inﬂuence each others’ ﬂames.

The revised model (see Figure 2) suggested in this study to assess the
effects of communication variables on media ﬂaming will face a bigger and
complex challenge in near future with the advent of outlets like blogs. Blogs, on-
line journals maintained by individuals, are not bound by traditional journalistic
norms, and therefore they might not follow a set pattern in their online content. In
other words, what it means is that these blogs might challenge the gate keeping

ﬂmction of traditional media, and refuse to conform with elite opinions (or

121

ﬂames), which seem to be widespread in the mainstream media. For instance, a
survey by Pew (2004a) during presidential campaign revealed that Democratic
presidential aspirant Howard Dean was successful in employing the Internet as an
effective campaign tool, and that a signiﬁcant number of his supporters visited his
campaign Web site and participated in on-line discussion on his political blog.
Therefore, in the near future, inter-disciplinary research with other ﬁelds such as
Information Technology might be necessitated to develop new theories to study
mass communication.

News gatherer resources should be included as a variable that might
inﬂuence ﬂames. The assumption here is that when the newsroom has adequate
editorial staff, then the reporters are under relatively less pressure to cover stories.
And when they have more time to cover stories, then they may try to contact
more sources for covering their stories, thereby bringing in a broader range of
ﬂames to their stories. On the other hand, fewer reporters in a newsroom would
put additional pressure on the small pool of journalists, in turn forcing them to
contact same sources again and again for sound bites and quotes, forcing them to
take a routine, formulaic approach to ﬂaming stories. With the use of the same
sources and a monotonous approach to their stories, journalists might end up
projecting the same kind of ﬂames in their news stories that sources need and
want.

The data in this study yielded moderate ﬁndings for the inﬂuence of
journalist gender on the ﬂaming of the Iraq issue. To gain a more elaborate

insight into effect of journalists’ gender on news ﬂaming, the revised model in

122

this study can be incorporated with in the gender model suggested by Rodgers and
Thorson (2003). In short, the gender model posits that men and women have
different workplace roles and behaviors because they are socialized differently
(for similar argument see Armstrong et al., 2006). The utility of this gender model
was established in Rodger and Thorson’s (2003) study, where they found that
while male and female journalists executed reporting tasks differently,
organizational factors such as circulation size and ratio of male-to-female editors,
mediated those differences. However, an early work by Breed (1955) on social
control in newsroom suggested that journalists can “bypass” newsroom
socialization by using their knowledge of speciﬁc beat or topic, by deciding
whom to interview and whom to ignore, which items to feature in the story and
which to bury, deciding on the tone of the story, and using their “star” status to
transgress policy. Based on Rodger and Thorson’s (2003) research ﬁndings in the
print medium, it is possible that organizational differences may also play a role in
gender representations of journalists in the electronic media (e.g., Armstrong et
al., 2006). For instance, this study’s data reﬂected that women journalists tend to
use more pro-Kerry ﬂames than their male counterparts. This ﬁnding might tie
out to the theory that women and men are socialized differently.

The revised model (see Figure 2) therefore incorporates male-to-female
ratio measure in the newsroom as it might reﬂect a signiﬁcant inﬂuence on news
media ﬂames. In other words, a relatively larger representation of women as
editors and reporters at news organizations might lead to ﬂaming issues and

events differently. A study by Craft and Wanta (2005), for instance, revealed that

123

newspapers with female editors often had less negative content than those with
male editors.

Similarly, to gain a deeper understanding of the inﬂuence of news gatherer
diversity, journalist race, but broadened into newsroom diversity, can also be
included in the revised model (see Figure 2). The data of this study, however,
suggested that journalist race was not a big factor in inﬂuencing the Iraq issue
ﬂames. But, while considering news gatherer diversity, journalist race cannot be
overlooked. For instance, as suggested by previous studies (e.g., Richardson &
Lancendorfer, 2004; Zeldes & F ico, 2005) non-white reporters tend to use more
non-white sources when compared to their white counterparts. Some studies have
also linked newsroom diversity to diversity in news ﬂaming (e.g., Gandy et al.,
1997). When compared with white journalists, non-white reporters might be able
to identify more with the non-white sources thereby using them more in their
stories. Therefore, by increasing diversity in sources non-white journalists might
ﬂame the stories differently their white counterparts. Such diversity can also be
operationalized in newsroom and management ratios.

Framing Power of Journalists grad Sources

A question that cannot be answered by the present study concerns who
really set the ﬂames—the journalists or the sources? In other words, did
journalists simply select the sources and ﬂames that ﬁt their preconceived notions
of how issues should be considered and the legitimacy of these sources and
ﬂames? To some extent, the answers to these questions might be explained by the

early work of Gieber and Johnson (1961) wherein they suggest that sources and

124

journalists stimulate and get stimulated by each other, in other words, they share a
symbiotic relationship. At the same time journalists themselves might inﬂuence
each others’ framing of stories, which stems ﬂom “pack journalism” (see Crouse,
1973)

Clearly, this relates to the relative power journalists have compared to
their sources when it comes to making news. The revised model incorporates
variables that relate to variation in power journalists and news organizations have
compared to their sources. Certainly, too, the answer to this question relates to the
media skill and focus of sources, and the monopoly they hold over information.

Journalists’ ﬂaming of issues or events in news stories depends on several
individual-level factors such as their background, socialization, mental maps of
the world, and gender. But then, newsroom resources such as number of reporters,
can also play a crucial role in how they exert their power in negotiating media
ﬂames. As noted above, if a newsroom has only a handful of reporters to cover a
large community then it is quite possible that one reporter might be working on
several stories in a day, placing that reporter more into the power of sources that
have their own “ﬂaming agenda.” And in the process of covering many stories in
a day, the reporter might simply end up transferring the intended ﬂame of
sources. On the other hand, if a newsroom has adequate editorial staff with
relatively less number of stories assigned to each reporter, then the chances of
their stories being more analytical with diverse ﬂames is more likely. This is
because reporters will have more time to plan and work on their stories and

contact diverse sources.

125

Having relatively more women and non-white reporters is also likely to be
reﬂected in presence of diverse ﬂames in the stories, whatever the pressure ﬂom
sources. This is because women and non-white reporters are socialized differently
and bring in their own unique perspectives to newsroom interactions of deﬁning,
sourcing and critiquing news stories. As women and non-white reporters are more
likely to use women and non-white sources, it is, then, very possible for their
stories to contain ﬂames different than the stories covered by their white, male
counterparts. Diversity in a newsroom, in other words, may lessen the chances
that reporter and source agreement in ﬂaming goes without challenge or question.

Most of the times journalists have to rely on hm sources and that
reliance may be more or less great. Certainly too, sources may consciously
attempt to sway media ﬂames, and how they interact with media will inﬂuence
their power to do so. Sources, especially skilled partisans, try to stay “on
message” in order to promote their own agenda and frames. This study showed
how disciplined such sources are at the national level in a high stakes election.
Although journalists rrright initially try to get a different opinion or view ﬂom the
source, when the source continues to reiterate the message, they too are forced to
relay that particular message. For instance, as indicated by this study’s data, the
Bush camp was successﬂrl in projecting Kerry as a ﬂip-ﬂopper who cannot lead
the nation in its war against terror. On the other hand, the Kerry camp kept
projecting the ﬂame of Bush administration’s failure in handling the Iraq issue.
While there were several deeper and complex dimensions tied to the Iraq issue,

the media ended up projecting only a few of those ﬂames sponsored by the

126

partisan sources. An explanation for such behavior of media is that journalists are
bound by deadlines, and in the rush to meet deadline they ﬁnd themselves
vulnerable to source ﬂames and in turn end up serving as mere conduits to those
ﬂames. At the same time, during events like Iraq war, they are inﬂuenced by
powerful images. For instance when a nation is engaged on a war, the image of
president might come to signify patriotic fervor, thereby making them to project
that image as it is their news coverage. In sum, it is a fact that in the media arena
there is a struggle for power between journalists and sources, and many a times it
is difﬁcult to decide a clear winner.

Moreover, it is also possible that journalists deliberately or inadvertently
use their power to distort reality in their own selection of ﬂames. As noted, this
study found that substantive ﬂames projected by partisans were ﬂequently
accompanied by the “political ﬁght” ﬂame. Such a ﬂame may be inappropriately
applied, debasing political debate and increasing public cynicism. Professional
steps journalistic organizations might take to prevent or correct such distortions
are discussed below.

Links Between Frmimd Other Theoretical Approaches

More generally, in line with Hallahan’s (1999) argument, social scientists
adopt a constructionist approach to emphasize that social issues are best
understood as problems that are built by claim makers (Best, 1995; Schneider,
1985). Elites and special interest groups vie for publicity in mass media by
generating support, forming coalitions, and using symbols to their advantage. In

other words, they are involved in a process of agenda-building (Cobb & Elder,

127

1972), which media may facilitate or hinder. Agenda-building involves nudging
issues ﬂom the public arena to the media arena and ﬁnally to the public policy
arena, where issues can get an ofﬁcial nod and validation (Manheim, 1987). But
then, only a few issues get public attention. Among variables that decide the
success of issue visibility is the limited carrying capacity of the system
(Hilgartrrer & Bosk, 1988), ﬂame enterprise and the efﬁciency of ﬂame
promotion by issue stakeholders (Gamson and Modigliani, 1989). This process
was conceptualized by Scheufele (2000) as ﬂame building (Hallahan, 1999).

The present study did not make an attempt to examine the process of
ﬂame building, especially as it involved active cooperation and opposition by
media. But future research can explicitly tie the ﬂaming theory with agenda-
building theory in order to show inﬂuence on media ﬂames. Further, ﬂaming
plays a crucial role in the process of agenda-building as interest groups try to
communicate with people either by directly or indirectly using the media. Thus, in
the later situation, social identiﬁcation theory posits that the objective is to signal
uninvolved group members about how they should think or act in regard to an
issue (Hallahan, 1999; Price, 1989).

For some time, researchers have called for examining the inﬂuence of
ﬂames on the news agenda (e.g., Aday, Cluverius, & Livingston, 2006). Linking
ﬂaming theory with the news agenda is likely to have more explanatory power for
assessing public opinion. Frarrring adds to the understanding of the relationship
between ﬂaming and agenda setting by showing how the former can give

dimensions and inﬂuence to the latter (for similar argument see Aday, Cluverius,

128

& Livingston, 2006). In other words, media ﬂames are speciﬁc about what to
think, not merely about what to think about, the basic power ascribed to
traditional agenda setting.

Moreover, as suggested by the ﬁrst Gulf War study of Allen et al. (1994),
combining the concept of priming and spiral of silence hypothesis with ﬂaming
theory could help show media’s role in making some opinions accepted and other
not. In other words, combining these processes would help in explaining how
media primed the Iraq issue to construct a reality that inﬂuenced the voters’
evaluation of this issue and also shaped the public debate as well. And in the
context of the spiral of silence, could media ﬂaming result in a prolonged
domination of particular public opinion?

While this study is primarily based on framing deﬁnitions of Entman
(1993), and Gamson and Modigliani (1987), the scope of ﬂrture research on the
Iraq issue ﬂaming can be broadened by borrowing ﬂom the ﬂaming deﬁnitions of
social movement. For instance, motivational ﬂaming involves a call to action as
well as the rationale for engaging in a corrective action. Frame alignment
represents efforts by various groups to foster coalitions and build relationships
(Snow, Rochford, Worden & Benford, 1986). Scholars of social movement
organizations have identiﬁed four strategies, “frame bridging (the linkage of
congruent with unconnected ﬂames), frame amplification (clariﬁcation and
crystallization of beliefs and values held by followers), frame extension (reaching
out to include other constituencies), and frame transformation (in which ﬂames

are altered in the wake of changing conditions),” (Hallahan, 1999, p. 219). These

129

ﬂaming concepts can be applied to examine the Iraq issue in a broader context of
American polity.

Ilapfigations for Joumali_sm

Resource Commitment

During a special and intense event like the presidential race, news outlets
are stretched in their resources, especially reporters. In addition to covering the
presidential race, the reporters are also expected to cover the routine news. If
journalists are assigned more stories than they can handle, then the quality of their
reporting suffers. In other words, they might just use a political ﬁght ﬂame to
emphasize horse race, which in turn would mean a story lacking the substantive
partisan ﬂames that deﬁne a problem or give a solution to it. Therefore, the news
organizations should increase their reporting staff in order to ensure that they are
not overworked.

While journalists work under immense pressure and have stiff deadlines to
meet, they can lessen this pressure by planning for their stories well in advance.
Planning in advance would help them to see through the ﬂames that their sources
are trying to promote and then take an informed decision to select and present
particular ﬂames in their stories. The editors should encourage the reporters to
diversify their source base so that stories could be covered ﬂom all possible
angles and include ﬂames that are considered important for shaping public
opinion. The newsrooms should also equip themselves with current technology in

order to increase efficiency and maintain quality in their news production process.

130

 

Journalists ' Socialization and Training

The data in this study suggest that the news media emphasized the horse
race coverage of the Iraq issue during the 2004 presidential race, commonly
linking it with substantive reporting. Journalists usually used the political ﬁght
ﬂame as a prelude to the partisan ﬂames. They would introduce a story by saying
how either of the candidates attacked each other over the Iraq issue or were using
the issue to some electoral advantage.

This kind of professional approach is disturbing, since by focusing more
on the conﬂict aspect, journalists might be failing in presenting a complete picture
of an issue or an event. Their efforts should be more on what sources have to say
about solving an issue rather than projecting it primarily as a political ﬁght. By
using formulaic, horse race approach, especially to cover election stories,
journalists rrright be doing a disservice to its audience. By focusing on the conﬂict
element in a campaign story it is likely for them to overlook what problem
deﬁnitions and solutions the presidential candidates have to offer.

This study found that some journalists at both networks and cable stations
served manifestly as conduits for transferring partisan Iraq issue ﬂames to the
public. Speciﬁcally, women journalists appeared to favor more pro-Kerry ﬂames
than pro-Bush ﬂames, which might be because women traditionally reﬂect a
Democratic bias. One way to address this issue is training reporters to add their
own analysis to news stories based on facts that they have gathered. And the other

way to make their stories less slanted is by encouraging them to go beyond

131

ofﬁcials sources by tapping into ordinary citizens in the community. Finally,
some standard of fairness and balance might be established and rationales applied.

As noted earlier, and consistent with prior research ﬁndings on campaign
coverage, television news served as conduits to the Iraq issue ﬂames sponsored
by the presidential candidates. One consequence of this is the minimal role *
accorded to non-partisans and other issue experts, who might have given broader
context and credibility to the discussion of issues. However, television news
cannot be totally blamed for this partisan focus on of the Iraq issue because, as
explained by Peer and Chestnut (2005), the thematic nature of television news
forces the broadcast reporters to repeat or cue into what their sources have said or
are going to say. Certainly news organizations must pay attention to the partisans.
But it is unclear why so little attention relatively is given to everyone else.

Fairness and Balance

With regard to fairness and balance in news, although most of the
reporters and anchors followed objective norms of journalism, at times their
biases for or against a particular presidential candidate would come through
clearly in their coverage. This is yet another reason for them to examine their
stories critically to see what kinds of ﬂames are being conveyed through them.
Because if a story’s fairness and balance favors a particular partisan source, then
the possibilities of ﬁnding that source’s ﬂames are higher, as also is indicated by
this study’s data. For instance, while covering an issue on which both the rival
candidates have said something, the journalists might give prominence to one

candidate over the other in their stories. Bias in a story’s fairness and balance

132

translates into its frames, as this study has shown. Research on news media
credibility (e.g., Fico, Richardson, & Edwards, 2004) has strongly suggested that
fairness and balance in news are tied to the news media credibility. One way
journalists could minimize bias in ﬂaming is by broadening their source base and
as far as possible using a neutral language.

Quality Control and Accountability

The professional world of journalism is a busy one, with reporters and
editors living by deadlines. But, the social responsibility that journalists have
might require editors taking that extra step to achieve and maintain high
standards. To ensure that the stories are not slanted or distorted, the editors can
check whether a reporter has used all possible sources necessary for the story and
not just the ofﬁcial sources. This way, ﬂames can be included not only ﬂom
partisan sources but also ﬂom ordinary citizens. Editors can also seek
participation ﬂom their own communities to give suggestions on the quality (or
lack thereof ) of their news coverage. By encouraging this kind of citizen
journalism, editors can improve their news coverage by getting helpful. inputs
ﬂom their own community. For instance, news media in the US. can form news
councils similar to the ones in Mexico. The El Norte newspaper group of Mexico
applied Reforma model since 1991 to increase community participation in
editorial-decision making by forming news councils (Chavez, 2005). Socially
active members of society and proactive readers such as those who write letters to
the editor were included in these news councils to discuss important issues in the

community and then help in the editorial-decision making process. Formation of

133

such news councils has improved the quality of news and credibility of news
organizations in Mexico (Chavez, 2005).

Further, editors can look at alternate media like blogs to see what they are
saying about various issues. There might still be an ongoing debate whether blogs
constitute a part of the mainstream media, but there is no dispute that they have
made their presence felt in the media landscape. Therefore, examining blogs for
discussions and opinions on issues might give additional insight and story ideas to
the editors. In sum, editors need to know whether stories are missing sources and
the ﬂames those sources project, whether or not stories are altered to
accommodate them. But news managements must be committed to gathering such
“alternate reality” information, at the least.

When it came to identifying non-partisan sources, this study questioned
whether the broadcast news media did a satisfactory job. While using non-partisan
sources, broadcast media usually failed to give their complete identiﬁcation,
thereby clouding source transparency. The mainstream media needs to emphasize
source transparency for two reasons. One, the norm of source transparency ties
into the concept of news media credibility. Second, in this digital age, with the
advent of blogs, the mainstream media cannot afford to be sloppy with sources, as
blogs not only identify sources but also provide hyperlinks to other sources for
cross reference. The upper-level management in the newsrooms should take steps
to maintain quality in their news production by maintaining adequate reporting

staff, training them for generating quality news stories by maintaining high ethical

134

standards, and making them credible and accountable for their news stories by
broadening their sources of information.

Implications for Research

Applying Generic/Issue Specific Frames

Both the revised model discussed above and the professional implications
of this study suggest directions for future research.

As the revised model suggests, ﬂrture studies can examine the issue-
speciﬁc ﬂames by linking them to the generic ﬂames, and then assessing ﬁndings
across studies at the generic level. For example, a comparative study of the news
coverage of the ﬁrst and second Gulf wars can be studied under broader generic
ﬂame of conﬂict, and then the coverage of both the wars can be examined for
issue speciﬁc ﬂames. Designing a study by anchoring issue-speciﬁc ﬂames to
generic ﬂames could facilitate cross-issue generalizations. In other words, any
issue in news media such as economy/job, abortion, same-sex marriage, and
health care can be selected and content analyzed for ﬂames in terms of generic
and issue-speciﬁc ﬂames. For instance, abortion, and same-sex marriage can be
examined under generic ﬂames of morality and human interest, and then within
those generic ﬂames these issues can be analyzed for presence of speciﬁc ﬂames.
Once content analysis studies standardize generic media ﬂames, then studies can
actually add cumulatively to our knowledge.

Varying Media Type and Frames

One of the major limitations of this study is that only six television news

channels were examined for media ﬂaming of the Iraq issue in 2004 presidential

135

race. Other important broadcast channels such as PBS were not included. PBS is
an important national news channel, which is known for its in-depth coverage of
issues and events. Therefore, future studies might include this channel.

A broader study including more media types such as elite vs. non-elite,
and traditional vs. blogs would help in getting a broader and better picture of
ﬂaming of the Iraq issue. This study looked at the national media, but a study of
local media would help to understand how this issue was being ﬂamed at local
level in different communities. Blogs have also made their presence felt in the
media landscape, which demands a scholarly study of this medium.

Varying News Gatherer Resources and Framing

Newsroom resources such as number of reporters in a news outlet plays an
important role in the quality if news coverage. If the newsroom has small number
of reporters covering many stories, then the increasing workload on them could
adversely affect their performance and in turn reﬂect ﬂaming in their news
stories. On the other hand, newsrooms having adequate number of reporters might
produce better stories as their editorial staff is relatively under less pressure.
Therefore, ﬁrture studies should consider newsroom resources as a variable to
assess its inﬂuence on framing.

Varying News Gatherer Diversity and Framing

To assess whether ﬂaming of issues and events are inﬂuenced by
newsroom diversity, studies should be conducted by considering journalist race as
a factor. The assumption here is that non-white journalists identify themselves

with non-white sources more than their white counterparts and tend to use them

136

more in their stories. So, to tease out the difference between white and non-white
journalists in their ﬂaming of the news future studies should be conducted in that
direction. Such studies are important because they have implications for
maintaining diversity in the newsroom.

Male/Female Ratio and Framing

Male-to—female ratio in a newsroom could also inﬂuence media ﬂames,
therefore future studies should be conducted in this direction. The assumption
here is that male and females are socialized differently and this can reﬂect in their
coverage of news. Also, representation of women in higher level positions in
newsroom is likely to inﬂuence how stories are covered and this can be another
line of research in near ﬁrture.

Content Analysis and its Limitations in Framing Research

The content analysis method has both promise and lirrritations in exploring
the process by which ﬂames enter the news. For example, it can be argued that
journalists inadvertently but importantly set news ﬂames by selectively giving
attention to some sources and ﬂames and ignoring others. An interesting study
might focus, for instance, on ﬂames set in candidate stump speeches and the
degree to which story ﬂames on those speeches rrrirror or distort those actually set
by the source.

Certainly, the content analysis method reaches the end of its utility when
trying to explain, for instance, why a male reporter might produce a story on a

candidate speech that more closely mirrors the candidate’s focus than did a story

137

covered by a female colleague. Here, interviews with the reporters and participant
observation are needed to explain the observed behavior differences.

Most importantly, to assess the inﬂuence of ﬂames, it must be asked
whether the ﬂaming made a difference in voters’ opinions and behaviors.
Certainly, content analysis combined with controlled experiments could help
illuminate any such inﬂuence. A good ﬁrst step would be to take the actual ﬂames
found in this content study as the experimental manipulations exposed to subjects
distinguished by factors such as partisanship, gender and race, and age and
education.

Socially Responsible Reporting and Scholarly Research

For a strong democracy, 3 free, fearless and vibrant news media is a must.
If news media merely tow the line of ofﬁcial and elite sources, they would be
failing in the watchdog function and risk just being a mouthpiece of a powerful
few, which is not desirable for a healthy democracy. If news media consistently
distort reality or bias their coverage, it is likely that the public will make poorer
public policy decisions.

In the present digital age, traditional journalism is facing interesting and
critical challenges ﬂom new media formats such as blogs. Citizens now have 24/7
access to news ﬂom various media outlets. Therefore, traditional news media
cannot take their role of informing people for granted.

One way to ensure credibility and audience loyalty is to maintain high
professional standards, and being aware of pitfalls such as slanting news stories in

favor of one or the other partisan group. This is the time when news organizations

138

have to do some soul searching to steer in a direction that ensures a well informed
citizenry, which in turn could result in a better society.

The way issues are ﬂamed during an electoral contest may later inform the
policies carried over into action. Scholarly research can assess how well news
media reﬂect different ways of looking at issues, and whether media ﬂames of
issues are distorted by inﬂuential sources or journalistic biases. If good journalism
is necessary for good political debate and public policy, then good media research
is crucial for a good journalism.

Scholarly research can help media organizations to monitor and assess
their performance as well as alerting the public to possible problems in media
news coverage. Scholarly research must theoretically and empirically assess how
journalism is projecting ﬂames of thought in society. Continued scholarly
examination of media ﬂaming of issues is, therefore, necessary, especially issue

ﬂaming during special and crucial events like presidential election campaigns.

139

APPENDIX 1

CONTENT ANALYSIS PROTOCOL—News Media Framing of the Iraq
Issue in the 2004 US. Presidential Campaign

 

 

 

 

 

Code Sheet

V1. Story Id.
V2. Station Id.
1=ABC 2=CBS 3=NBC 4=CNN 5=FOX

6=MSNBC
V3. Story Date (mm/dd)
V4. Story Length (Seconds)
V5. Story Type

1=anchor lead-in 4=anchor vo/ sot 5=anchor in-house
interview

=anchor reader 6=reporter package

3=anchor vo 7=reporter live
V6. Journalist Gender

1=man 2=woman
V7. Journalist Race

1= white 2= non-white (Identify speciﬁc

Source Number Seconds Given
Bush Partisans v 8. v9.
Kerry Partisans v10. v11.
Partisan Story Domination
Favors Favors Bal.
NA
Republicans Democrats

V12. Partisan Side Speaking First 1 2
V13. Time Given Sides 1 2 3
V14. Visual l 2 3
9
V15. Direct Quotes 1 2 3
9

140

V16. Partisan Domination 1 2 3

(v12-v15)

V 17. Non-ﬂame story seconds
V18. Dominant Iraq Issue Frame set by the Republican(s)

 

Time in sec.
l= Fighting against terrorism ﬂame
2= Fighting for ﬂeedom and democracy ﬂame
3= Kerry incompetence ﬂame
4= Bush competence ﬂame
5= Conﬂict ﬂame
6= Political ﬁght ﬂame
7= Consensus ﬂame
8= Policy failure/responsibility ﬂame
9= Economic consequence ﬂame
10= Bush incompetence ﬂame
11= Kerry competence ﬂame
12= Other ﬂame

 

V19. Dominant Iraq Issue Frame set by the Democrat(s)

 

Time in sec.
1= Fighting against terrorism ﬂame
2= Fighting for ﬂeedom and democracy ﬂame
3= Kerry incompetence ﬂame
4= Bush competence ﬂame
5= Conﬂict ﬂame
6= Political ﬁght ﬂame
7= Consensus ﬂame
8= Policy failure/responsibility ﬂame
9= Economic consequence ﬂame
10: Bush incompetence ﬂame
11= Kerry competence ﬂame
12= Other frame

 

NON-PARTISAN SOURCE DOMINATION

Source Number Seconds Given
Experts v 20. v21.

Ordinary Citizens v22. v23.

141

Journalists v24. v25.

V 26. Dominant Iraq Issue Frame set by non-partisan source(s)

 

Time in sec.
1= Fighting against terrorism ﬂame
2= Fighting for ﬂeedom and democracy ﬂame
3= Kerry incompetence ﬂame
4= Bush competence ﬂame
5= Conﬂict ﬂame
6: Political ﬁght ﬂame
7= Consensus ﬂame
8= Policy failure/responsibility ﬂame
9= Economic consequence ﬂame
10= Bush incompetence ﬂame
ll= Kerry competence ﬂame
12= Other ﬂame

 

lllllllllll

V27. Dominant Iraq Issue Frame set by Journalists

 

Time in sec.
1= Fighting against terrorism ﬂame
2= Fighting for ﬂeedom and democracy ﬂame
3= Kerry incompetence ﬂame
4= Bush competence ﬂame
5= Conﬂict ﬂame
6= Political ﬁght ﬂame
7= Consensus ﬂame
8: Policy failure/responsibility ﬂame
9= Economic consequence ﬂame
10= Bush incompetence ﬂame
11= Kerry competence ﬂame
12= Other ﬂame

 

V28. Overall Dominant Frame

llllllllllll

V29. Sponsor of the ﬂame (circle one answer below).

1. Republicans 2. Democrats 3. Non-partisan 4. Journalist

142

CODING PROTOCOL—News Media Framing of the Iraq Issue in the 2004
US. Presidential Campaign

The main newscast of six network and cable channels will be taped ﬂom Labor
Day to the election day. A researcher will review tapes and mark any Iraq issue
story that is covered in the context of 2004 presidential race.

All marked stories will be numbered and coded by station, date, time and position
in the tape. An Iraq issue story is one that has been covered in the context of the
presidential race. Speciﬁcally, the story deals with the presidential candidates,
their parties, supporters, and others debating, discussing, commenting and talking
about the Iraq issue in the context of presidential election. Although, many
political issues are covered by the news media during the study period, for the
purpose of this study only the Iraq issue will be considered.

The Iraq Issue Stories:

The Iraq issue stories must contain at least one assertion on the Iraq war/issue
made in the context of election by a partisan or nonpartisan source or by the
journalist. A partisan source may be the candidate, staff, family or party
supporters, or simply a citizen who supports one of the candidates for president. A
single partisan assertion establishes the partisanship of the source for all election
coverage. A source who has not made a partisan assertion in any story is
considered a non-partisan source. Assertions lacking any attribution to a source
are considered the journalist’s.

An assertion ﬂom one or more of these partisan sources may be quoted or
paraphrased. It may be by a source on air or attributed to some source by the
reporter or anchor. A sign of graphic by itself that appears in a story does not
constitute such an assertion.

An Iraq issue story in the context of election is a story whose topic is Iraq and
issues related with Iraq. The transition from one Iraq issue story to another
Iraq issue story is determined by change in election story topics. However, a
change in stories is considered to have taken place when one journalist such
as an anchor hands off a story to a reporter, even if the topic is the same.
Protocol Definitions:

V1. Story ID: Each story gets a unique identifying number

V2. Station ID: Each station gets a unique identifying number

V3. Story Date: Each story is dated by month and day

V4. Story Length: Story length is measured in seconds with a stop watch

143

V5. Story Type

l= Anchor Lead-in: This type of story consists of an anchor’s introduction to a
Reporter
Package (see below) or a Reporter Live (see below) that follows
immediately. The reporter MUST follow. If no Reporter Package or
reporter Live follows, see codes 2-4 below.
2= Anchor Reader: This type of story consists TOTALLY of the anchor’s own
report. Any graphics or pictures in the studio are NOT considered visual material.

3= Anchor VO: This type of story consists of the anchor’s own report
ACCOMPANIED by visual material, but NOT by any other voices (voice that
constitute background “noise” are not considered.

4: Anchor VO/SOT (and interview): This type of story consists of the anchor’s

own
report ACCOMPANIED by the visual material AND other voices.
Includes sources interviewed in the studio by the anchor.

5= Anchor Interview: Includes interview along with NO video material

6= Reporter Package: This type of material consists of a reporter’s own report that
MUST BE accompanied by visual material and other voices. A change in
reporters is considered to demark a NEW PACKAGE.

7= Reporter Live: The anchor will introduce a reporter “live” ﬂom the ﬁeld. The
anchor may interact with the reporter, or only the reporter will be seen or heard
following the anchor’s introduction. A Reporter Live ACCOMPANIED by taped
or other material ﬂom that reporter is considered to be a Reporter Package (see 5
above).

V6. Journalist Gender: Record journalist’s gender.

V7. Journalist Race: Record reporter race.
1= White
Skin color: pale reddish, white to olive brown
Hair color and texture: light blond to dark brown, and ﬁne

2= Non-white sources
Skin color: safﬂon to yellow, or reddish brown to brown-black
Hair color and texture: dark, straight and coarse.

V8, V10. Determine how many Republican and Democratic sources make
partisan Assertions

144

V9, V11. Determine how many total seconds Republicans and Democratic
sources get.

a. Candidate or partisan assertions are measured using a stopwatch.

b. Assertions are indicated by the presence of a named partisan linked
with a verb of attribution such as “said,” “asserted,” etc., or a verb
denoting state of mind such as “feels,” believe,” etc.

0. The total assertions times of opponents that are within three seconds of
one another are considered to be identical in length. The recorded
length will be the average of the assertions of the two sides.

V12: Partisan Side Speaking First: Identify the side speaking ﬁrst in the election
story

V13: Time Given Sides: Determine ﬂom V11 and V 13 what side time favors

V14: Visual: Identify if one or both sides appear in a visual of any kind in the
news story

V15: Direct Quotes: Identify if one or both sides are directly quoted through voice
broadcast

V16: Partisan Domination: Count the number of components in V12-V15
favoring Republicans and Democrats.
a. If more of the relevant components favor the Republicans, score the
story as favoring the Republicans
b. If more of the relevant components favor the Democrats, score the
story as favoring the Democrats
c. If an equal number of components favor the Republicans and
Democrats, score the story as Balanced.

V17. Non-ﬂame story seconds: With a stop watch determine total number of
seconds devoted to a non-ﬂaming material in a news story. Non-ﬂaming material
includes straight/ factual reporting such as news anchor reporting of a Bush trip to
Ohio or Kerry in Florida. For example, a journalist may report that the Bin Laden
tapes surfaced during the presidential campaign and both the candidates
responded to the incident. Here the reporter is stating a fact and is not trying to
ﬂame how each of the presidential candidates is responding. In other words, non-
ﬂaming seconds are wherein the content is factual and is not being ﬂamed in a
speciﬁc way.

V18. Iraq Issue Frame set by the Republican(s)

Time in sec.
l= Fighting against terrorism ﬂame
2= Fighting for ﬂeedom and democracy ﬂame
3= Kerry incompetence ﬂame

 

145

4= Bush competence ﬂame

5= Conﬂict ﬂame

6= Political ﬁght ﬂame

7= Consensus ﬂame

8= Policy failure/responsibility ﬂame
9= Economic consequence ﬂame
10= Bush incompetence ﬂame

11= Kerry competence ﬂame

12= Other ﬂame

FRAMING OF THE IRAQ ISSUE (V18 to V19, V26 to V27)

DEFINITION: Framing is deﬁned as the way sources or journalists describe or
present the Iraq issue in a news story. Sources or journalists will select some
aspect of the Iraq issue and make it more salient in the news story, in such a way
as to promote a particular problem deﬁnition, causal interpretation, moral
evaluation, and/ or treatment recommendation.

Rule for Coders: Journalists will be considered as the ﬂamers of the story ONLY
when there are no sources or verb of attribution such as said, felt, and hoped to
sources in what they say/report.

CODING: The coders should look for how the sources are ﬂaming (describing or
presenting) the Iraq issue or people associated with the Iraq policy in a news
story. For instance, President Bush may describe the Iraq war is an attempt to
defeat terrorism, which indicates that the Iraq issue is being framed as war
against terrorism. With a stop watch record the number of seconds being devoted
to this ﬂame (description).

A story might have multiple sources projecting multiple ﬂames. In such a
situation, record number of seconds devoted to each ﬂame by different category
of sources (Republicans, Democrats, Experts, and Journalists). Add up the total
ﬂaming seconds and then within this ﬂaming time ﬁnd out if any deﬁned ﬂame
accounts for 50% of the ﬂame material time to determine the Dominant Frame.

For instance, in a news story out of 60 seconds of ﬂaming material time if
30 seconds are devoted to Fighting for Freedom and Democracy Frame and 15
seconds each to Political Fight Frame, and Responsibility/Policy Failure Frame,
then the Fighting for Freedom and Democracy Frame will be determined as
dominant ﬂame as it accounts for 50% of the total ﬂaming material time.

Note for Coders: When the ﬂaming material does not fall under any of the

deﬁned categories or is confusing then code it as the Other Frame, see the
deﬁnition below.

146

1= FIGHTING AGAINST TERRORISM FRAME

Here the focus is on equating the Iraq war/issue with the ﬁght against terrorism.
The source (s) will explicitly project the ﬂame that the Iraq war/issue is a ﬁght
against terrorism. For example, President Bush may emphasize that Iraq was axis
of evil under Saddam Hussein’s regime and that world is better off without him
being in power.

KEY WORD/PHRA SE EXAMPLE:

Fight against terrorism, axis of evil, world better of without Saddam, supporting
terrorists, weapons of mass destruction, protecting America from terrorists, etc.

2= FIGHTING FOR FREEDOM AND DEMOCRACY FRAME

Here the focus is on equating the Iraq war/issue with bringing ﬂeedom and
democracy to Iraqi people. The source (s) will explicitly project the ﬂame that
Iraqi people have been liberated ﬂom the dictatorship of Saddam Hussein. Efforts
are being made to establish a democratically elected government in Iraq.

KEY WORD/PHRA SE EXAMPLE:

Democracy, liberation, freedom, rescuing Iraqi people from tyranny of Saddam,
bringing freedom and democracy to Iraqi people, etc.

3= KERRY INCOMPETENCE FRAME

Here the focus is that Kerry is a ﬂip-ﬂopper. Source (s) would ﬂame Kerry as
indecisive and that he changes his issue position on the Iraq war, which makes
him unﬁt to lead the nation, speciﬁcally handling the Iraq issue.

KEY W 0RD/PHRA SE EXAMPLE:

F lip-flapper, indecisive, Kerry does not make a good leader, etc.

4= BUSH COMPETENCE FRAME

Here the focus is that Bush is a better commander-in—chief to lead the nation in
right direction. Bush can do a better job than Kerry at the domestic ﬂont as well
as in foreign policy such as the war in Iraq.

KEY WORD/PHRA SE EXAMPLE:

I will be a better leader, I will be a better commander-in-chieﬂ 1 can lead the
nation in right direction, etc.

5= CONFLICT FRAME

Here the focus is on Bush and Kerry claims in the context of the terrorism and
Iraq issue that they would provide a safer America. To be considered a conﬂict
ﬂame, sources should be explicitly referring to the broader international conﬂict
rather than to the election. The Republican and Democrat partisans will argue
respectively that they can ﬁght the terrorism better.

KEY WORD/PHRA SE EXAMPLE:

I will make a better leader, I will handle Iraq situation better, attack, etc.

147

6= POLITICAL FIGHT FRAME

Here sources explicitly identify that one or both sides, Republicans and
Democrats, are using the Iraq issue to secure some political or electoral gain, as a
tactical strategy or to be considered a political ﬁght ﬂame.

KEY W ORB/PHRASE EXAMPLE:

Shoring up support, wedge issue, political gain on support, winning voter
strategy, etc.

7= CONSENSUS FRAME

Here the focus is on Bush and Kerry agreeing or expressing consensus on some
aspect of the Iraq issue. TO BE CONSIDRED A CONSENSUS STORY BOTH
THE PARTISAN SOURCES SHOULD BE PRESENT OR ATTRIBUTED IN
THE STORY. For instance, when Bin Laden tapes threatening America asking to
withdraw troops ﬂom Iraq were released during presidential campaign, Bush and
Kerry agreed that they will never give in to the demands of terrorists. In such
situations Bush and Kerry emphasized that America remains united, and is not
Republican or Democrat when it comes to tackling terrorism.

KEY W 0RD/PHRA SE EMMPLE:

We are united to ﬁght against terrorism, we will not give in to the threats of
terrorists, we are united and not Republican or Democrat when it comes to
fighting terrorism, etc.

8= RESPONSIBILITY/POLICY FAILURE FRAME

Here the focus is that the Bush administration is responsible for a mess in Iraq and
that the Iraq policy is a failure. The focus is on Bush administration, and not Bush
as an individual. For instance, Kerry might argue that the Bush administration
took a hasty decision of invading Iraq and it reﬂects a poor thinking and planning
on the part of current government.

KEY WORD/PHRASE EXAMPLE: Responsible, reason, policy failure, hasty
decision, etc.

9= ECONOMIC CONSEQUENCE FRAME

Here the focus is on economic consequence of the Iraq war. For example, Kerry
might point out that the ill-planned Iraq war has cost over $200 billion, which
could have been used for education and generating jobs.

KEY WORD/PHRA SE EXAMPLE:

Economic cost of the Iraq war, billions of dollars have been wasted on the war,
war is causing economic burden on the country, etc.

10= BUSH INCOMPETEN CE FRAME

Here the focus is on Bush’s incompetence to handle the Iraq issue. The focus is on
Bush as an individual. For instance, Kerry might allege that President Bush
misled the country by claiming that Iraq stored weapons of mass destruction and
took the nation to the war without taking the citizens into conﬁdence.

KEY WORD/PHRA SE EXAMPLE:

Cannot handle the Iraq issue, misled, arrogant, poor planner, etc.

148

ll= KERRY COMPETENCE FRAME

Frame projected here is that Kerry will keep focus on terrorism, not on Iraq. Kerry
will be smarter than Bush. Kerry will get allies, world support to hunt down and
kill terrorists.

KEY WORD/PHRASE EXAMPLE:

Competence, allies, winning international supporters, global test, etc.

12= OTHER FRAME

Use the other ﬂame when the ﬂames projected by the sources don’t fall into any
of the deﬁned categories above. Or when the projected ﬂame (s) cannot be clearly
established.

V19. Iraq Issue Frame set by the Democrat(s)

"l
5
(D
5.
m
8

1= Fighting against terrorism ﬂame
2= Fighting for ﬂeedom and democracy ﬂame
3= Kerry incompetence ﬂame

4= Bush competence ﬂame

5= Conﬂict ﬂame

6= Political ﬁght ﬂame

7= Consensus ﬂame

8= Policy failure/responsibility ﬂame
9= Economic consequence ﬂame
10= Bush incompetence ﬂame

11= Kerry competence ﬂame

12= Other ﬂame

 

 

(Follow the same ﬂaming rules and deﬁnitions as under V18)

V20, V22, V24. Determine how many experts, ordinary citizens and journalist
sources make non-partisan Assertions. Non-partisan expert is anyone who is
explicitly mentioned as expert in the news story (not afﬁliated to any political
party), can be a pollster, scholar, political analyst. Ordinary citizen is any
common man interviewed in the story. Journalists are individuals covering news
as reporters or anchors.

V21, V23, V25: Count the time given to non-partisan sources

V26. Iraq Issue Frame set by the non-partisan source(s)

Time in sec.
1= Fighting against terrorism ﬂame
2= Fighting for ﬂeedom and democracy ﬂame
3= Kerry incompetence ﬂame
4: Bush competence ﬂame

 

149

5= Conﬂict ﬂame

6= Political ﬁght ﬂame

7= Consensus ﬂame

8= Policy failure/responsibility ﬂame
9= Economic consequence ﬂame
10= Bush incompetence ﬂame

ll= Kerry competence ﬂame

12= Other ﬂame

(Follow the same ﬂaming rules and deﬁnitions as under V18)

V27 . Iraq Issue Frame set by Journalists

Time in sec.
1= Fighting against terrorism ﬂame
2= Fighting for ﬂeedom and democracy ﬂame
3= Kerry incompetence ﬂame
4= Bush competence ﬂame
5= Conﬂict ﬂame
6= Political ﬁght ﬂame
7= Consensus ﬂame
8= Policy failure/responsibility ﬂame
9= Economic consequence ﬂame
10= Bush incompetence ﬂame
11: Kerry competence ﬂame
12= Other ﬂame

 

(Follow the same ﬂaming rules and deﬁnitions as under V18)

V28. Overall Dominant Frame

A story might have multiple sources projecting multiple ﬂames. In such a
situation, record number of seconds devoted to each ﬂame by different category
of sources (Republicans, Democrats, Non-partisans, and Journalists). Add up the
total ﬂaming seconds and then within this ﬂaming time ﬁnd out if any deﬁned
ﬂame accounts for 50% of the ﬂame material time to determine the Dominant
Frame.

For instance, in a news story out of 60 seconds of ﬂaming material time if
30 seconds are devoted to Fighting for Freedom and Democracy Frame and 15
seconds each to Political Fight Frame, and Responsibility/Policy Failure Frame,
then the Fighting for Freedom and Democracy Frame will be determined as
dominant ﬂame as it accounts for 50% of the total ﬂaming material time.

V29. Sponsor of the dominant ﬂame (circle one answer below).

1. Republicans 2. Democrats 3. Non-partisan 4. Journalist

150

Appendix II
Frame Type Examples
Eggples of Pro-Bush Frames
1. Fighting Against Terrorism Frame

President GEORGE W. BUSH: We're chasing down these killers overseas so we
don't have to face them here at home. (CBS, Evening News, September 8, 2004)

President GEORGE W. BUSH: Saddam Hussein would still be in power. He
would control all those weapons and explosives and could have shared them with
our terrorist enemies. (NBC Nightly News, 2004).

 

Pres. BUSH: I wasn't happy when we found out there wasn't weapons, but
Saddam Hussein was a unique threat and the world is better off without him in
power. (NBC Nightly News, October 9, 2004).

 

President GEORGE W. BUSH: If we stop ﬁghting the terrorists in Iraq, they
would be ﬂee to plot and plan attacks elsewhere, in America and other ﬂee
nations. (CBS Evening News, September 23, 2004).

2. Fighting for Freedom and Democracy Frame

President GEORGE W. BUSH: They [Iraqis] are going to have elections in
January of 2005. The world is becoming a better place because ﬂeedom is on the
march. (CBS Evening News, September 17, 2004).

3. Kerry Incompetence Frame

President GEORGE W. BUSH: During the course of this campaign, the fellow
I'm running against has probably had about eight positions in on Iraq. Let me be
clear: Mixed signals are the wrong signals to send to our troops in the ﬁeld, the

Iraqi people, to our allies and most of all to our enemies. (CBS Evening News,
September 16, 2004).

 

President GEORGE W. BUSH: He apparently woke up this morning and has now
decided, no, we should not have invaded Iraq, after just last month saying he still
would have voted for force, even knowing everything we know today. (NBC
Nightly News, September 20, 2004).

 

151

KELLY O'DONNELL (Reporter): The trouble for Kerry is that he did vote to
authorize the war and even today defended that vote, turning the blame on the
president, saying Mr. Bush failed to use it properly. Kerry aides say they believe
they can move beyond that now with a more pointed critique on the worsening
conditions in Iraq and by challenging the president on what should be done next.
The president ﬁred back ﬂom New Hampshire accusing Kerry of "twisting in the
wind."

President GEORGE W. BUSH: He apparently woke up this morning and has now
decided, no, we should not have invaded Iraq, after just last month saying he still
would have voted for force, even knowing everything we know today.

O'DONNELL: The Bush campaign characterizes Kerry's view as "retreat and
defea "...

Sen. KERRY: Our troops have served...
O'DONNELL: ...while Kerry claims the administration's choices and troubling

conditions on the ground are reason enough for voters to choose change. Kelly
O'Donnell, NBC News, New York. (NBC Nightly News, September 20, 2004).

 

PRESIDENT GEORGE W. BUSH

Having gone back and forth so many times, the Senator ﬂom Massachusetts has
now ﬂip-ﬂopped his way to a dangerous position. My opponent, my opponent
ﬁnally has settled on a strategy, a strategy of retreat.

TERRY MORAN (Reporter)

(Voice Over) On ﬁghting terrorism, Mr. Bush accused Kerry of having, quote, a
"September 10th attitude" and, quote, "a policy of weakness."

PRESIDENT GEORGE W. BUSH

He says that preemptive action is unwise not only against regimes, but even
against terrorist organizations. (ABC World News Tonight, October 18, 2004).

 

WOLF BLITZER (Anchor): Iraq has emerged as the central issue in the election.

Without mentioning his Democratic challenger by name, the president pointedly
used the Allawi visit to go after John Kerry's supposed ﬂip-ﬂops on the Iraq war.

"I believe a leader must be consistent and clear and not change positions when
times get tough. And the times have been hard. These are hard times. But I

152

understand what mixed messages do. You can embolden an enemy by sending
mixed messages," President Bush said. (CNN Wolf Blitzer Reports, September
23, 2004).

4. Bush Competence

PRESIDENT GEORGE W. BUSH, UNITED STATES

And in this time of choosing, I want all Americans to know you can count on me
to ﬁght our enemies and defend our ﬂeedom. ABC World News Tonight, October
18,2004)

 

PRESIDENT GEORGE W. BUSH, UNITED STATES

In the ﬁnal four days of this historic campaign, I'm taking my vision of a more
hopeful America directly to the people of this country. (ABC World News
Tonight, October 29, 2004)

Examples of Neutral Frames
5. Conﬂict

President GEORGE W. BUSH: Let me make this very clear. Americans will not
be intimidated or inﬂuenced by an enemy of our country.

Senator JOHN KERRY (Democrat, Massachusetts; Presidential Nominee): As
Americans, we are absolutely united in our determination to hunt down and
destroy Osama bin Laden and the terrorists. (CBS Evening News, October 29,
2004).

6. Political Fight
TOM BROKAW, anchor:

President Bush and John Kerry are at war over the war in Iraq. And while the
president ties it to terrorism, the issue that serves him best, Senator Kerry opened
a new ﬂont today linking the war to the American economy and social issues.
This on a day when new questions have surfaced about the president's service in
the National Guard. We're covering both campaigns tonight. We'll begin with
NBC's Kelly O'Donnell and the latest Kerry strategy. (NBC Evening News,
September 8, 2004).

 

153

TERRY MORAN (Reporter)

(Voice Over) Before a raucous partisan crowd here, Mr. Bush launched a scathing
attack on Senator Kerry's record and policies on national security issues. On Iraq,
a harsh accusation. (ABC World News Tonight, October 18, 2004).

 

TERRY MORAN (Reporter), ABC NEWS

(Voice Over) This morning in Manchester, New Hampshire, after weeks of
slashing away at Senator John Kerry, President Bush offered an argument for his
reelection, without once mentioning his rival. (ABC World News Tonight,
October 29, 2004)

 

DEAN REYNOLDS (Reporter)

(Voice Over) On Iraq, Kerry seized on reports that the former US army
commander there, General Ricardo Sanchez, had complained a year ago of
inadequate supplies for things like tanks and helicopters to ﬁght the insurgents.
And Kerry noted other former ofﬁcials have belatedly admitted they asked for
more troops, as well, to no avail. (ABC World News Tonight, October 14, 2004).

 

7. Consensus Frame
DAN RATHER, anchor:

With just four days left in the presidential campaign, Osama bin Laden suddenly
dropped himself right in the middle of it with a videotaped message to the people
of the United States. The ﬂrgitive al-Qaida leader admits for the ﬁrst time that he,
indeed, ordered the September 11th attack on America. He lays out his reasons for
it, and he threatens another attack. This tape tends to confrrrn that bin Laden is
alive and well, safely somewhere.

President Bush responded by saying the United States will not be intimidated.
Senator Kerry said the country is united in its determination to hunt bin Laden
down. CBS' Bob Orr begins our coverage. (CBS Evening News, October 29,
2004)

Examples of Pro-Kerry Frames

8. Policy F ailure/Responsibility Frame

Senator JOHN KERRY: The president's policy in Iraq has not strengthened our
national security; it has weakened it.

154

O'DONNELL (Reporter): Today, Kerry went further than he ever has before,
Using the equation, consider everything known today, Kerry now says the war
was wrong and labeled a series of the president's judgments as a "colossal
failure."

Sen. KERRY: If we know there was no imminent threat, no weapons of mass
destruction, no ties to al-Qaeda, the United States should have invaded Iraq? My
answer is resoundingly ‘No!’

O'DONNELL: Kerry repeatedly mentioned Osama bin Laden and said invading
Iraq was a profound diversion ﬂom that real enemy.

Sen. KERRY: We have traded a dictator for chaos that has left America less
secure.

O'DONNELL: The trouble for Kerry is that he did vote to authorize the war and
even today defended that vote, turning the blame on the president, saying Mr.
Bush failed to use it properly. Kerry aides say they believe they can move beyond
that now with a more pointed critique on the worsening conditions in Iraq and by
challenging the president on what should be done next. (NBC Nightly News,
September 20, 2004).

 

DEAN REYNOLDS (Reporter)

(Voice Over) quickly turned serious, at a news conference that included a
scathing review of the president's speech to the UN.

SENATOR JOHN KERRY

The president of the United States stood before a stony-faced body and barely
talked about the realities, at all, of Iraq.

DEAN REYNOLDS

(Voice Over) Kerry cast Bush as a failed leader.

SENATOR JOHN KERRY

The president really has no credibility at this point. And he has no credibility with
ﬁgﬁign leaders who hear him come before them and talk as if everything is going
DEAN REYNOLDS

(Voice Over) Once again, though, Kerry was at pains to explain why if he thinks

Iraq is such a mess, he voted to give Bush authority to go to war. But this time, he
boiled it down to just two sentences.

155

SENATOR JOHN KERRY

My vote was a vote to do this the right way. And had I been president, we would
have done this the right way.

DEAN REYNOLDS

(Voice Over) How, asked Kerry, could anyone have known that the Bush war
plan, as he described it, would ignore expert advice, leave Iraq's borders and
ammunition dumps unguarded, or be so slow to rebuild the country?

SENATOR JOHN KERRY

This has been incompetently handled, mismanaged every step of the way. (ABC
World News Tonight, September 21 , 2004).

9. Economic Consequence Frame

AXELROD (Reporter): The war alone is not a big winner for Kerry. Many voters
approve of the president's handling of it, plus, Kerry voted for it. But Kerry's now
taken it a step further, linking the cost of war in Iraq with a money shortage for
problems at home.

Sen. KERRY: $200 billion that we're not investing in education, in health care,
job creation here at home.

AXELROD: Again and again, Kerry tied the $200 billion ﬁgure to home-ﬂont
issues.

Sen. KERRY: I call this course a catastrophic choice that has cost us $200 billion
because we went it alone. We've paid an even more unbearable price in young
American lives. (CBS Evening News, September 8, 2004).

 

KELLY O'DONNELL reporting:

Good evening, Tom. Tonight a more aggressive John Kerry is trying to reﬂame
the discussion on the president's signature issue. Kerry's new buzzword,
"choices," urging voters to see the costs of the war in Iraq differently.

Today John Kerry's retooled line of attack attempts to fuse the issue he struggles

with most, Iraq, with one topic that advisers believe gives Kerry his best shot at
unseating the president: the economy.

156

Senator JOHN KERRY: And while we're spending that $200 billion in Iraq—-that's
to this date; it will go on--eight million Americans are looking for work here in
America.

O'DONNELL: Kerry chose Cincinnati, the same spot where Mr. Bush made his
case two years ago for toppling Saddam...

President GEORGE W. BUSH: (From ﬁle footage) The threat comes ﬂom Iraq.

O'DONNELL: ...and chose to repeat the president's own words of assurance then
about careful planning, all to link the growing costs of war to unmet needs at
home.

Sen. KERRY: Two hundred billion dollars for going it alone in Iraq. That's the
wrong choice. That's the wrong direction. And that's the wrong leadership for
America.

O'DONNELL: A theme repeated in Kerry's latest ad.

Unidentiﬁed Man: (From Kerry ad) Two hundred billion dollars for Iraq. In
America, lost jobs and rising health-care costs. (NBC Nightly News, September 8,
2004).

10.Bush Incompetence Frame

TERRY MORAN (Reporter)

(Voice Over) The president's bottom line, containment of Saddam Hussein was no
longer an option. Democrats were having none of that.

SENATOR JOHN F. KERRY, DEMOCRATIC PRESIDENTIAL NOMINEE
The president this morning was in absolute full spin mode.
TERRY MORAN

(Voice Over) Senator Kerry said the president was grasping at straws, denying the
reality of Iraq and looking for new reasons to justify the war.

SENATOR JOHN F. KERRY

Ladies and gentlemen, the president of the United States and the vice president of
the United States may well be the last two people on the planet who won't face the
truth about Iraq. (ABC World News Tonight, October 7, 2004)

 

157

SENATOR JOHN F. KERRY, DEMOCRATIC PRESIDENTIAL NOMINEE

Despite the president's arrogant boasting that he's done everything right in Iraq
and that he's made no mistakes, the truth is beginning to come out, and it's
beginning to catch up with him. (ABC World News Tonight, October 18, 2004).

 

11. Kerry Competence Frame

WOLF BLITZER (Anchor): On the campaign trail in Columbus, Ohio, Kerry shot
right back saying, "I want victory. I want to win. And I have a better plan to win
than George Bush does. The president says that things are getting better in Iraq
and we must just stay the course. Well, I disagree. They're not getting better and
we need to change the course to protect our troops and to win." (CNN Wolf
Blitzer Reports, September 23, 2004)

 

ROBERTS: But is it a ﬂip-ﬂop? Kerry has consistently said holding Saddam
accountable was and remains the right thing to do. And he's been just as
consistent in his opposition to the way President Bush went to war.

Sen. KERRY: It's not I would have done just one thing differently in Iraq; I would
have done everything differently in Iraq. (CBS Evening News, September 7,
2004).

Other Frames
TERRY MORAN (Reporter)

(Off Camera) In a statement, Bremer says he believes current troop levels are
sufﬁcient, and that his opinion that more were needed related only to the early
stages of the occupation. He also says he strongly supports President Bush's
reelection... (ABC World News Tonight, October 5, 2004).

E_x_ample of No Fra_m_e

HEADLINE: THE KERRY CAMPAIGN DEMOCRATIC HOPEFUL
ENERGIZED

PETER JENNINGS (Anchor), ABC NEWS
(Off Camera) Senator Kerry, as we know, has been talking about Iraq for a long
time and the president's leadership there. And he's been talking about this

ammunition dump for a couple of days in a very intense way. Here's ABC's Dean
Reynolds. (ABC World News Tonight, October 27, 2004).

158

References

ABC. (2004). Damage control, president on the defensive. ABC World News
Tonight, October 7.

ABC. (2004). Bush “out of touch,” Kerry criticizes Bush. ABC World News
Tonight. October 1 8.

Aday, S., Cluverius, J ., & Livingston, S. (2005). As goes the statue, so goes the
war: The emergence of the victory ﬂame in television coverage of the Iraq
war. Journal of Broadcasting & Electronic Media, 49, 314-331.

Allen, B., O’Loughlin, P., Jasperson, A., & Sullivan, J. L. (1994). The media and
the Gulf War: Framing, priming, and the spiral of silence. Polity, 27, 255-
284.

Andsager, J. L., & Powers, A. (1999). Social or economic concerns: How news
and women’s magazines ﬂamed breast cancer in the 1990s Journalism
and Mass Communication Quarterly, 76, 531—550.

Armstrong, L. C., Wood, L. M., & Nelson, M. (2006). Female news professionals
in local and national broadcast news during build up to the Iraq war.
Journal of Broadcasting & Electronic Media, 50, 78-94.

Bae, H. S. (2000). Product differentiation in national TV newscasts: A
comparison of the cable all-news networks and the broadcast networks.
Journal of Broadcasting & Electronic Media, 44, 62- 77.

Ball-Rokeach, S., & Rokeach, M. (1987). The ﬂrture study of public opinion: A
symposium. Public Opinion Quarterly, 51, 184-185.

Baumgartner, F. R., & Jones, B. D. (1993). Agendas and instability in American
politics. Chicago: University of Chicago Press.

Bennett, W. L. (1990). Toward a theory of press state relations in the United
States. Journal of Communication, 40 (2), 103-125.

Bennett, W. L. (1994). The news about foreign policy. In W.L. Bennett & D.L.
Paletz (Eds), Taken by storm: The media, public opinion, and US. foreign
policy in the Gulf War (pp. 12-42). Chicago: University of Chicago Press.

Bennett, W. L. (1995). News: The politics of illusion. New York: Longman.

Bennett, W. L. (1996). The governing crisis: Media, money and marketing in

American elections. New York: Toward a theory of press state relations in
the United States. Journal of Communication, 40 (2), 103-125.

159

Bennett, W. L., & Manheim, J. B. (1993). Taking the public by storm:
Information, cuing, and the democratic process in the Gulf conﬂict.
Political Communication, 1 0, 33 1-551 .

Benson, R. (2004). Bringing the sociology of media back in. Political
Communication, 21 , 275-292.

Best, J. (1995). Images of Issues: T ypijying contemporary social problems (2nd
ed.). New York: Aldine de Gruyter.

Breed, W. (1955). Social control in the newsroom: A functional analysis. Social
Forces, 33 (4), 326-335.

Brewer, P. R. (2002). Framing, value words, and citizens’ explanations of their
issue opinions. Political Communication, 19, 303-316.

Callaghan, K., & Schnell, F. (2001). Assessing the democratic debate: How the
news media ﬂame elite policy discourse. Political Communication, 18,
1 83-212.

Cappella, J. N., & Jamieson, K. H. (1997). Spiral of cynicism: The press and the
public good. New York: Oxford University Press.

Carter, 8., Fico, F ., & McCabe, A. J. (2002). Partisan and structural balance in
local television election coverage. Journalism & Mass Communication
Quarterly, 79, 41-53.

CBS. (2004). Bush and Kerry campaigns continue to argue over their positions
regarding the war in Iraq. CBS Evening News, September 8.

CBS. (2004). John Kerry and George Bush address the war in Iraq in campaign
speeches. CBS Evening News, September 16.

CBS. (2004). Presidential candidates argue over national security. CBS Evening
News, October 20.

Center for Media and Public Affairs. (2003). The media go to war: TV news
coverage of the war in Iraq. Media Monitor, 1 7(2).

Chavez, M. (2005). News agendas and community participation: A Mexican
model for change. Orrline resource accessed ﬂom
http://uclaccc.ucla.edu/articlesﬁrrticle-_m_anuelclmrvezreport.htm

Chyi, I. H., & McCombs, M. (2004). Media salience and the process of ﬂaming:
Coverage of the columbine school shootings. Journalism and Mass
Communication Quarterly, 81, 22-35.

160

Cigler, A. J ., & Loorrris, B. A. (1995). Contemporary interest group politics: More
than “more of the same.” In A. J. Cigler & B. A. Loomis (Eds), Interest
group politics (pp. 393-407). Washington, DC: CQ Press.

CNN. (2004). Bush, Allawi united on Iraq. CNN Wolf Blitzer Reports. September
23.

CNN. (2004). WolfBlitzer Reports. October 23.

Cobb, R. W., & Elder, C. D. (1983). Participation in American Politics: The
dynamics of agenda building. Baltimore: Johns Hopkins University Press.

Cohen, B. (1963). The press and foreign policy. Princeton, NJ: Princeton
University Press.

Congressional Quarterly Weekly Report. (2003). 22 March.

Cook, T. E. (1998). Governing with the news: The news media as a political
institution. Chicago: University of Chicago Press.

Craft, 8., & Wanta, W. (2004). Women in the newsroom: Inﬂuences of female
editors and reporters on the news agenda. Journalism & Mass
Communication Quarterly, 81 , 124-138.

Crouse, T. (1973). The Boys on the Bus. New York: Ballantine.

D’Alessio, D. & Allen, M. (2000). Media bias in presidential elections: A meta-
analysis. Journal of Communication, 50, 133-156.

D’Angelo, P. (2002). News ﬂaming as a multiparadigmatic research program: A
response to Entman. Journal of Communication, 52 (4), 870-888.

Dahinden, U. (2002). Biotechnology in Switzerland—ﬂames in a heated debate.
Science Communication, 24 (2), 184-197.

Damore, F. D. (2005). Issue convergence in presidential campaigns. Political
Behavior, 2 7, 71-97.

Danchev, A., & MacMillan, J. (2005). The Iraq War and Democratic Politics.
London; New York: Routledge.

de Vreese, C. H. (2004). The effects of ﬂames in political television news on

issue interpretation and ﬂame salience. Journalism and Mass
Communication Quarterly, 81 , 36-52.

161

de Vreese, C. H., Peter, J ., & Semetko, H. A. (2001). Framing politics at the
launch of the Euro: A cross-national comparative study of ﬂames in the
news. Political Communication, 18, 107-122.

Delli Carpini, M. X., & Keeter., S. (1991). Stability and change in the US.
public’s knowledge of politics. Public Opinion Quarterly, 55, 583-612.

Dimitrova, D. V., Kaid, L. L., Williams, A. P., & Trammell, K.D. (2005). War on
the Web: The immediate news ﬂaming of Gulf War II. Harvard
International Journal of Press/Politics, 10, 22—44.

Donohue, A. G., Tichenor, J. P., & Olien, N. C. (1995). A guard dog perspective
on the role of the media. Journal of Communication, 45, 115-132.

Druckrnan, J. N. (2005). Media matter: how newspapers and television news
cover campaigns and inﬂuence voters. Political Communication, 22, 463-
481.

Dumbrell, J. (2005). Bush’s war: the Iraq conﬂict and American democracy. In
Alex Danchev & John MacMillan (Eds), The Iraq War and Democratic
Politics (pp. 35-46). London; New York: Routledge.

Edelrnan, M. J. (1993). Contestable categories and public opinion. Political
Communication, 10, 231-242.

Entman, R. M. (1989). Democracy without citizens: media and the decay of
American politics. New York: Oxford University Press.

Entman, R. M. (1991). Framing US coverage of international news: Contrasts in
narratives of the KAL and Iran air incidents. Journal of Communication,
41 (4), 6-27.

Entman, R. M. (1993). Framing: Toward clariﬁcation of a ﬂactured paradigm.
Journal of Communication, 43 (4), 51-58.

Entman, R. M. (2004). Projections of power: Framing news, public opinion, and
US. foreign policy. Chicago : University of Chicago Press.

Entman, R. M, & Page, I. B. (1994). The news before the storm: The Iraq war
debate and the limits to media independence. In W. L. Bennett & D. L.
Paltez (Eds), Taken by Storm: The Media, Public Opinion, and US.
Foreign Policy in the Gulf War (pp. 82-104). Chicago: University of
Chicago Press.

162

Entman, R. M., & Rojecki, A. (1993). Freezing out the public: Elite and media
ﬂaming of the US. anti-nuclear movement. Political Communication, 10,
155-173.

Esser, F ., & D’Angelo, P. (2003). Framing the press and publicity process.
American Behavioral Scientist, 46, 617-641.

F ico, F., & Balog, O. (2003). Partisan sources receive more space in conﬂict
issues. Newspaper Research Journal, 24, 22-35.

Fico, F., Clogston, J ., & Pizante, G. (1988). Inﬂuence of candidate party and
incumbency on news coverage of the 1984 elections in Michigan dailies.
Journalism Quarterly, 65, 709-713.

F ico, F., & Cote, W. (2002). Partisan and structural balance of election stories on
the 1998 govemor’s race in Michigan. Mass Communication & Society, 5,
167-184.

Fico, F., & Freedman, E. (2001). Setting the news story agenda: Candidates and
commentators in news coverage of a govemor’s race. Journalism & Mass
Communication Quarterly, 78, 437-449.

Fico, F., Ku, L., & Soffm, S. (1994). Fairness, balance of newspaper coverage of
US. in Gulf War. Newspaper Research Journal, 15, 30-43.

Fico, F ., Richardson, J., & Edwards, S. (2004). Inﬂuence of story structure on
perceived story bias and news organization credibility. Mass
Communication & Society, 7, 301-318.

Fico, F., & Soffm, S. (1995). Fairness and balance of selected newspaper
coverage of controversial national, state, and local issues. Journalism &
Mass Communication Quarterly, 72, 621-633.

F ico, F ., Zeldes, G. A., & Diddi, A. (2004). Partisan and structural balance of
local television election coverage of incumbent and open gubernatorial
elections, Journalism & Mass Communication Quarterly, 81, 897-910.

Freedman, E., & Fico., F. (2004). Whither the experts? Newspaper use of horse
race and issue experts in coverage of open governors’ races in 2002.
Journalism & Mass Communication Quarterly, 81, 498-510.

Freedman, E., & Fico, F. (2005). Male and female sources in newspaper coverage

of male and female candidates in open races for governor in 2002. Mass
Communication & Society, 8, 257-272.

163

Gallagher, M. (2001). Gender Setting: News agendas for media monitoring and
advocacy. London: Zen Books.

Gamson, WA. (1992). Talking politics. New York: Cambridge University Press.

Gamson, WA. (1996). Media discourse as a framing resource. In A. N. Cigler
(Ed.), The psychology of political communication (pp. 111-132). Ann
Arbor: University of Michigan Press.

Gamson, W. A., & Modigliani, A. (1987). The changing culture of afﬁrmative
action. In Richar D. Braungart (Ed), Research in Political Sociology
(Vol. 3, pp. 137-177). Greenwich, Conn: JAI.

Gamson, W. A., & Modigliani, A. (1989). Media discourse and public opinion on
nuclear power: A constructionist approach. American Journal of
Sociology, 95,1-37.

Gandy, O. H., Jr., Kopp, K., Hands, T., Frazer, K., & Phillips, D. (1997). Race
and risk: factor affecting the framing of stories about inequality,
discrimination, and just plain bad luck. Public Opinion Quarterly, 61, 158-
182.

Gans, H. (1980). Deciding what’s news: A study of CBS Evening News, NBC
Nightly News, Newsweek and Time . New York: Pantheon.

Gartner, S. S., Segura, M. S., & Wilkening, M. (1997). All politics are local: local
losses and individual attitudes toward the Vietnam War. Journal of
Conﬂict Resolution, 41 , 669-694.

Ghanem, S. (1997). Filling in the tapestry: the second level of agenda setting. In

Maxwell McCombs, Donald L. Shaw, & David Weaver (Eds.),

Communication and democracy: Exploring the intellectual frontiers in
agenda-setting theory (pp. 3-14). Mahwah, NJ: Lawrence Erlbaum
Associates.

Gidengil, E., & Everitt, J. (2003). Talking tough: gender and reported speech in
campaign news coverage. Political Communication, 20, 209-232.

Gieber, W., & Johnson, W. (1961). The city hall beat: a study of reporter and
source roles. Journalism Quarterly, 38, 289-297.

Gitlin, T. (1980). The whole world is watching. Berkeley: University of California
Press.

Goffman, E. (1974). Frame analysis. Boston: Northeastern University Press.

164

Graber, DA. (1993). Media power in politics (3rd ed.). Washington, DC: CQ
Press.

Hallahan, K. (1999). Seven models of framing: Implications for public relations.
Journal of Public Relations Research, 11, 205-242.

Hallin, DC. (1992). Sound bite news: Television coverage of elections, 1968-
1988. Journal of Communication, 42 (2), 5-24.

Hallin, D., & Mancini, P. (1984). Speaking of the president: Political structures
and representational form in the US. and Italian television news. Theory
& Society, 13, 829-850.

Hiebert, RE. (2003). Public relations and propaganda in framing the Iraq war: a
preliminary review. Public Relations Review, 29, 243-255.

Hilgartner, S., & Bosk, C. L. (1988). The rise and fall of social problems: A
public arenas model. American Journal of Sociology, 94 (I), 53-78.

Holbert, R. L., Benoit, W. L., Hansen, G. J ., & Wen, W. (2002). The role of
communication in the formation of an issue-based citizenry.
Communication Monographs, 69 (4), 296-310.

Hutcheson, J ., Domke, D., Billeaudeaux, A., & Garland, P. (2004). US. national
identity, political elites, and a patriotic press following September 11.
Political Communication, 21, 27—50.

Iyengar, S. (1991). Is anyone responsible? How television frames political issues.
Chicago: University of Chicago Press.

Iyengar, S., & Kinder, D. R. (1987). News that matters: Television and American
opinion. Chicago: University of Chicago Press.

Jacoby, G. W. (2000). Issue framing and public opinion on government spending.
American Journal of Political Science, 44, 750—767.

J anowitz, M. (1975). Professional models in journalism. The gatekeeper and the
advocate. Journalism Quarterly, 52, 918-626, 662.

Jasperson, A., Shah, D. V., Watts, M., Faber, R., & Fan, DP. (1998). Framing the

public agenda: Media effects on the importance of the federal budget
deﬁcit. Political Communication, 15, 205-224.

165

Johnson, J. T., Braima, A. M., & Sothirajah, J. (2000). Measure for Measure: The
relationship between different broadcast types, formats measures, and
political behaviors and cognitions. Journal of Broadcasting & Electronic
Media, 44, 43-61.

Katovsky, B., & Carlson, T. (2003). Embedded: the media at war in Iraq.
Guilford, CT: Lyons Press.

Kerschbaumer, K. (2003, August 11). RTNDA, UNITY call for diversity summit.
Broadcasting & Cable, 133 (32), 10-11.

Kinder, D. , & Sanders, L. (1996). Divided by color: Racial politics and
democratic ideals. Chicago: University of Chicago Press.

Kerbel, M. R., Apee, S., & Ross, M. H. (2000). PBS ain’t so different: public
broadcasting, election frames, and democratic empowerment. Harvard

International Journal of Press/Politics, 5 (4), 8-32.

Lawry, D. T., & Shidler, J. A. (1995). The sound bites, the biters, and the bitten:
An analysis of network TV bias in campaign ‘92. Journalism& Mass
Communication Quarterly, 72, 33-43.

Lawry, D. T., & Shidler, J. A. (1998). The sound bites, the biters, and the bitten:
A two-campaign test of the anti-incumbent bias hypothesis in network TV
news. Journalism & Mass Communication Quarterly, 75, 719-729.

Lippmann, W. (1920). Liberty and the news. New York: Harcourt, Brace, &
Hone.

Love, B. R., & Fico, F. (2006). Framing by reporters, partisan and non-partisan
sources in newspaper coverage of US. senate races in 2004. Paper
accepted for presentation at the annual conference of the International
Communication Association, Dresden, Germany.

Luther, C. A., & Miller, M. M. (2005). Framing of the 2003 U.S.-Iraq war
demonstrations: An analysis of news and partisan texts. Journalism &
Mass Communication Quarterly, 82, 78-96.

Manheim, J. D. (1987). A model of agenda dynamics. In M. McLaughlin (Ed.),
Communication Yearbook (Vol. 10, pp. 449-516). Thousand Oaks, CA:
Sage.

Martin, C. R., & Oshagan, H. (1997). Disciplining the workforce: The news

media frame a General Motors plant closing. Communication Research,
24, 669-697.

166

McChesney, R. (2002). September 11th in the mind of American journalism. In B.
Zelizer and S. Allen (Eds.), Journalism after September 11’”, (pp.9l-100).
New York: Routledge.

McCombs, M. F., Llamas, J. P., Lopez-Escobar, E., & Rey, F. (1997). Candidate
images in Spanish elections: Second-level agenda-setting effects.
Journalism & Mass Communication Quarterly, 74, 703-717.

McCombs, M. F., Shaw, D. L., & Weaver, D. L. (1997). Communication and
Democracy: Exploring the Intellectual Frontiers in A genda-Setting
Theory. Marwah, NJ: Lawrence Erlbaum.

McLeod, D.M., & Detenber, B. H. (1999). Framing effects of television news
coverage of social protest. Journal of Communication, 49(3), 3-23.

McLeod, D. M. & Hertog, J. K. (1992). The manufacture of public opinion by
reporters: Informal cues for public perception of protest groups. Discourse
and Society, 3, 259.

Media Institute. (1983). CNN vs. the networks: Is more news better news?
Washington DC: Media Institute.

Mendelsohn, M. (1993). Television news frames in 1988 Canadian election.
Canadian Journal of Communication, 18, 149-171.

Meyer, D.S. (1995). Framing national security: Elite public discourse on nuclear
weapons during the Cold War. Political Communication, 12, 173-192.

Mills, K. (1997). What difference do women journalists make? In . P. Norris
(Ed.), Women, Media and Politics (pp. 77-98). Oxford: Oxford University

Press.

Monroe, B. (2003). Newsroom diversity: Truth vs. ﬁction. Nieman Reports, 5 7
(3), 29.

Morreale, J. (1991). A New Beginning: A Textual Frame Analysis of the Political
Campaign Film. Albany: State University of New York.

Muller, E. J. (1973). War, presidents, and public opinion. New York: John Wiley.

NBC. (2004). Issue of war in Iraq takes center stage in presidential campaign.
NBC Nightly News. September 8.

NBC. (2004). Presidential campaigns battle Iraq policies. NBC Nightly News.
September 20.

167

Nelson, T., Clawson, R. A., & Oxley, Z. M. (1997). Media framing of civil
liberties conﬂict and its effect on tolerance. American Political Science
Review, 91, 567-583.

Neuman, W. R., Just, M. R., & Crigler, A. N. (1992). Common knowledge: news
and the construction of political meaning. Chicago: University of Chicago
Press.

Norris, P. (1995). The restless searchlight: Network news framing of post-coldwar
world. Political Communication, 12, 357-370.

Norris, P., Kern, M., & Just, M. (2003). Framing terrorism: the news media, the
government, and the public. New York : Routledge.

Page, I. B., Shapiro, Y. R., & Dempsey, R. G. (1987). What moves public
opinion? American Political Science Review, 81 , 23-43.

Pan, Z., & Kosicki, G. M. (1993). Framing analysis: An approach to news
discourse. Political Communication, 10, 55-76.

Parmelee, J. (2002). Presidential primary Videocassettes: How candidates in the
2000 US. presidential primary elections framed their early campaigns.
Political Communication, 19, 317-331.

Patterson, TE. (1993). Out of Order. New York: A. Knopf.

Patterson, TE. (1998). Political roles of journalists. In D. Graber, D. McQuail, &
P. Norris (Eds), The politics of news, the news of politics (pp. 17-32).
Washington, DC: CQ Press.

Patterson, TE. (2004). Iraq issue propels election interest: Young adults affected
most. Joan Shorenstein Center on the Press, Politics and Public Policy.
Online resource retrieved from
http://www.vgnishingvoter.org/Releases/release09l 704.shtml

Patterson, T. E., & Donsbach, W. (1996). News decisions: Journalists as partisan
actors. Political Communication, 13, 455—468.

Peer, L., Chestnut, B. (1995). Deciphering media independence: The Gulf War
debate in television and newspaper news. Political Communication, 12,
81-95.

Pew Research Center for the People and the Press. (2004). Voters Liked

Campaign 2004, But T 00 Much 'Mud-Slinging.‘ Online resource retrieved
from http://people-press.org/reports/disnlav3hp3?ReportID=233

168

Pew Research Center for the People and the Press (2004a). Cable and Internet
Loom Large in Fragmented Political News Universe. Online resource
retrieved from http://people-press.org/repons/displjayphw?ReoortID=200

Project for Excellence in Journalism. (2002). Return to normalcy? How the media
have covered war on terrorism. Online resource retrieved from
http://www.joum_alism.org/resources/research/reports/normalcy

Project for Excellence in Journalism. (2006). The state of news media 2006: an
annual report on American journalism. Online resource retrieved from
http://www.stateoﬂhenewsmedig.com/2006/index.asp

Price, V., Tewksbury, D., & Powers, E. (1997). Switching trains of thought: The
impact of news frames on readers’ cognitive responses. Communication
Research, 34, 481-506.

Reese, S. D., & Buckalwe, B. (1995). The materialism of local television: The
routine framing of the Persian Gulf War. Critical Studies in Mass
Communication, 12, 40-59.

Rhee, J. W. (1997). Strategy and issue frames in election coverage: A social
cognitive account of framing effects. Journal of Communication, 47 (3),
26—48.

Riffe, D., Lacy, S., & Fico, F. (2005). Analyzing media messages: Using
quantitative content analysis in research. Mahwah, NJ: Lawrence
Erlbaum.

Richardson, D. J ., & Lancendorfer, K. (2004). Framing afﬁrmative action: the
inﬂuence of race on newspaper editorial responses to the University of
Michigan cases. Harvard International Journal of Press/Politics, 9 (4),
74-94.

Rodgers, 8., & Thorson, E. (2003). A socialization perspective on male and
female reporting. Journal of Communication, 53, 658-678.

Ross, D. S., & Bantimaroudis, P. (2006). Frame shifts and catastrophic events: the
attacks of September 11, 2001, and New York Time’s portrayals of Arafat
and Sharon. Mass Communication & Society, 9, 85-101.

Ross, K., & Srebemy , A. (2000). Women in the House: media representation of
British politicians. In A. Srebemy & L. van Zoonen (Eds), Gender,
Politics and Communication (pp. 79-99). Cresskill, NJ: Hampton Press.

Ross, J., & Taylor, P. (1992). The new news v. the old news. Washington, DC:
Brookings Institution.

169

Ryan, C., Carragee, M. K., & Meinhofer, W. (2001). Framing, the news media,
and collective action. Journal of Broadcasting & Electronic Media, 45 ,
175-1 82.

Sahr, R. (1993). Credenrialing the experts: The climate of opinion and journalist
selection of sources in domestic and foreign policy. In R]. Spitzer (Ed.),
Media and public policy (pp. 153-170). Westport, CT: Praeger.

Scheufele, D. A.(1999). Framing as a theory of media effects. Journal of
Communication, 49(1), 103-122.

Scheufele, D. A. (2000). Agenda-setting, priming, and framing revisited: Another
look at cognitive effects of political communication. Mass Communication
& Society, 3, 297-316.

Scheufele, D. A., Nisbet, M. C., & Ostman, R. E. (2005). September 11 news
coverage, public opinion, and support for civil liberties. Mass

Communication & Society, 8, 197-218.

Schneider, J. W. (1985). Social problems theory: the constructionist view.
American Review of Sociology, I I, 209-229.

Schneider, W. (2003). War has its reasons. National Journal, 22 March.
Schneider, W. (2003a). Still asking, why now? National Journal, 8 February.

Schudson, M. (1983). The news media and the democratic process. (Wayne
Resource Paper.) New York: Aspen Institute for Humanistic Studies.

Sears, 0. D. (1993). Symbolic politics: A socio-psychological analysis. In Shanto
Iyengar & William J. McGuire (Eds), Explorations in Political
Psychology (pp. 113-149). Durham, NC: Duke University Press.

Semetko, H. A., Blumler, J. G., Gurevitch, M., & Weaver, D. H. (1991). The
formation of campaign agendas: A comparative analysis of party and
media roles in recent American and British elections. Hillsdale, NJ:
Erlbaum.

Semetko, H. A., & Valkenburg, P. M. (2000). Framing European politics: A
content analysis of press and television news. Journal of Communication,
50(2), 93-109.

Shah, D., Domke, D., & Wackman, D. B. (1996). “To thine own self be true”:

Values, framing, and voter decision-making strategies. Communication
Research, 23, 509-560.

170

Shen, F. (2004). Effects of news frames and schemas on individuals’ issue
interpretations and attitudes. Journalism and Mass Communication

Quarterly, 81, 400-416.

Shoemaker, P. J ., & Reese, S. D. (1996). Mediating the message: Theories of
inﬂuences on mass media content. White Plains, NY. : Longman.

Simon, A., & Xenos, M. (2000). Media framing and effective public deliberation.
Political Communication, 17, 363-376.

Snow, D. A., Rochford, E. B., Jr., Worden, S. K., & Benford, R. K. (1986). Frame
alignment process, micromobilization, and movement participation.
American Socilogical Review, 51, 464-481.

Sparrow, B.H. (1999). Uncertain guardians: The news media as a political
institution. Baltimore: Johns Hopkins University Press.

Stempel, G. H. (1988). Topic and story choice of ﬁve network newscasts.
Journalism Quarterly, 65, 750-752.

Sutcliffe, J. B., Lee, F. M., & Soderlund, C. W. (2005). Women and crisis
reporting: Television news coverage of political crises in the Caribbean.
Harvard International Journal of Press/Politics, 10, 99-124.

Trimble, L., & Sampert, S. (2004). Who’s in the game? The framing of the
Canadian election 2000 by the Globe and Mail and the National Post.
Canadian Journal of Political Science, 3 7, 51-71.

Tuchman, G. (1978). Making news: A study in the construction of reality. New
York Free Press.

Valkenburg, P. M., Semetko, H. A., & de Vreese, C. H. (1999). The effects of
news frames on readers’ thoughts and recall. Communication Research,
26, 550-569.

van Zoonen, L. (1994). Feminist Media Studies. London: Sage.

Wanta, W. (1997). The Public and the National Agenda: How People Learn
about Important Issues. Mahwah, NJ: Lawrence Erlbaum.

Weaver, D. H., & Wilhoit, G. C. (1986). The American journalist. A portrait of
U.S. news people and their work (2nd ed.). Bloomington: Indiana
University Press.

171

Wicks, H. R., & Walker, C. D. (1993). Differences between CNN and the
broadcast networks in the live war coverage. In Bradley S. Greenberg &
Walter Gantz (Eds), Desert Storm and the Mass Media (pp. 99-112).
Cresskill, N.J.: Hampton Press.

Williams, B. A. (2003). The new media environment , Internet chatrooms, and
public opinion discourse after 9/11. In Daya Kishan Thussu & Des
Freedman (Eds), War and the Media: reporting conﬂict 24/7 (pp. 176—
189). London: Sage.

Zaller, J. (1990). Experimental tests of the question-answering model of the mass
survey response. Ann Arbor. University of Michigan.

Zaller, J ., & Chiu, D. (1996). Govemment’s little helper: U.S. press coverage of
foreign policy crises, 1945-1991. Political Communication, 13, 385-405.

Zeldes, G. A., & F ico, F. (2005). Race and gender: An analysis of sources and
reporters in the networks’ coverage of the 2000 presidential campaign.
Mass Communication & Society, 8, 373-385.

Zelizer, B., Allan, S. (2002). Journalism after September 11. New York:
Routledge.

172

   

I“glitygjinjjjjuigijju