The Messenger Overwhelming the Message: the Impact of Ideological Cues on Information Acquisition, Counterarguing, and Perceptions of Media Bias Joel F

Florida State University Libraries

Electronic Theses, Treatises and Dissertations The Graduate School

2007 The Messenger Overwhelming the Message: The Impact of Ideological Cues on Information Acquisition, Counterarguing, and Perceptions of Media Bias Joel F. Turner

Follow this and additional works at the FSU Digital Library. For more information, please contact [email protected]

THE FLORIDA STATE UNIVERSITY

COLLEGE OF SOCIAL SCIENCES

THE MESSENGER OVERWHELMING THE MESSAGE: THE IMPACT OF IDEOLOGICAL CUES ON INFORMATION ACQUISITION, COUNTERARGUING, AND PERCEPTIONS OF MEDIA BIAS

JOEL F. TURNER, JR.

A Dissertation Submitted to the Department of Political Science in partial fulfillment of the requirements for the degree of Doctor of Philosophy

Degree Awarded: Spring Semester 2007

The members of the Committee approve the dissertation of Joel F. Turner, Jr. defended on March 27, 2007.

______Robert Jackson Professor Directing Dissertation

______Tim Salmon Outside Committee Member

______Jeffery Mondak Committee Member

______Charles Barrilleaux Committee Member

______Cherie Maestas Committee Member

Approved:

______Dale L. Smith Chair, Department of Political Science

______David Rasmussen Dean, College of Social Sciences

The Office of Graduate Studies has verified and approved the above named committee members.

ii ACKNOWLEDGEMENTS

This dissertation benefits greatly from the insightful comments of those gracious enough to share their wisdom. I would like to most sincerely thank Jeff Mondak for his guidance in this undertaking. His commitment to this project has been unwavering and his timely feedback was critical to the completion of this dissertation. I am thankful to Bob Jackson, whose methodological suggestions and vast knowledge of the English language resulted in a dissertation that was far more polished than it otherwise would have been. I am grateful to Charles Barrilleaux, who has always been willing to share wisdom and expertise on matters ranging from this project to the job market to college football. I would like to thank Cherie Maestas for her helpful comments, as well as for bringing me on board for other exciting research endeavors. I am also indebted to Tim Salmon for his insight on this study as well as for agreeing to serve on a committee outside of his department. I would also like to thank Rod Lewis for his assistance, as well as for serving as a constant source of support during my graduate school experience. I am grateful to Karen Halperin for her assistance in both conducting this experiment and collecting the data used in this analysis. I would also like to thank Mary Schneider, our department’s academic coordinator, for always having the right answer to any question we had, and basically ensuring that the department functioned properly on a daily basis. I would also like to extend a special thank you to Phil Streetman and WSST-TV 55 in Cordele, GA for their assistance in creating the videotapes used in this study. Without your help, getting this project off the ground would not have been possible. Finally, I wish to thank my family - Candice, Mom, and Dad - for their encouragement and patience in pursuit of this goal. I could not have reached this point without their support. I especially wish to thank my beautiful wife Candice for her unwavering confidence in my ability to complete this endeavor despite setbacks along the way. Your patience and willingness to sacrifice on many things in order to allow me to pursue my goals has not gone unnoticed, and I only hope that over time I will be able to repay your love and support.

iii TABLE OF CONTENTS

List of Tables……………………………………………………………………………...v List of Figures…………………………………………………………………………….vi Abstract…………………………………………………………………………………..vii

INTRODUCTION………………………………………………………………………...1

RESEARCH METHODOLOGY………………………………………………………...14

THE MESSENGER OVERWHELMING THE MESSAGE: IDEOLOGICAL CUES AND PERCEPTIONS OF BIAS IN TELEVISION NEWS……………………..30

MEMORABLE ENCOUNTERS: IDEOLOGY, INFORMATION ACQUISITION, AND TELEVISION NEWS…………………………………………..46

I HEAR WHAT YOU’RE SAYING, I JUST DON’T BELIEVE YOU: COUNTERARGUING AND DISSONANT MEDIA SOURCES………………………68

CONCLUSION…………………………………………………………………………..81

APPENDIX A……………………………………………………………………………88

APPENDIX B……………………………………………………………………………96

APPENDIX C……………………………………………………………………………97

APPENDIX D……………………………………………………………………………98

REFERENCES…………………………………………………………………………105

BIOGRAPHICAL SKETCH…………………………………………………………...115

iv LIST OF TABLES

Table 2.1…………………………………………………………………………………25

Table 2.2…………………………………………………………………………………26

Table 2.3…………………………………………………………………………………27

Table 2.4…………………………………………………………………………………28

Table 2.5…………………………………………………………………………………29

Table 3.1…………………………………………………………………………………43

Table 3.2…………………………………………………………………………………44

Table 4.1…………………………………………………………………………………63

Table 4.2…………………………………………………………………………………64

Table 4.3…………………………………………………………………………………65

Table 5.1…………………………………………………………………………………79

v LIST OF FIGURES

Figure 3.1………………………………………………………………………………...45

Figure 4.1………………………………………………………………………………...66

Figure 4.2………………………………………………………………………………...67

Figure 5.1………………………………………………………………………………...80

vi ABSTRACT

Public opinion surveys have demonstrated that many Americans perceive ideological bias in television news. For many individuals, the two networks that epitomize this bias are Fox News Channel and CNN, the networks that many believe represent the conservative and liberal viewpoints, respectively. These perceptions persist despite the best efforts of those in the academic community, who have failed to uncover systematic evidence of the existence of ideological bias in the television news media. In this study, I argue that attaching the “CNN” and “Fox News Channel” labels to news stories essentially functions as an ideological signal, cueing viewers to the ideological tone of the coverage they are about to watch. In addition, I argue that these perceptions of ideological bias can have significant consequences for the ability of viewers to properly process the information presented on the news broadcast. An experimental design was utilized to investigate these questions, and the results provide support for my expectations: the “Fox News Channel” and “CNN” labels do appear to function as a signal regarding the ideological nature of the coverage, and this signal does significantly influence the ability of viewers to acquire and retain the information presented on the broadcast.

vii

CHAPTER ONE

INTRODUCTION

Public opinion surveys have consistently demonstrated that a large portion of the citizenry perceive an ideological bias within television news media (Pew 1997; American Society of Newspaper Editors 1998; Gallup 2003). For many Americans the two television news outlets that epitomize this bias are CNN and Fox News Channel, which according to Nielsen1 ratings are the two most popular twenty-four hour news networks. Although surprisingly little academic research has been done on these specific networks, various polling estimates have demonstrated that roughly one-third of the American public perceives Fox News Channel as being overtly conservative, and a comparable portion views CNN as being liberally biased.2 Given the well-established role of the news media as political informant to the American public, it is important to evaluate two questions: first, why do individuals perceive ideological bias in the news media, and, second, does this perceived ideological bias influence how people process television news? There is a distinct possibility that the widespread view among citizens that Fox News Channel and CNN are ideologically biased poses a substantial obstacle to informing the American public about politics. Perceiving a report from a specific news outlet as biased could raise cognitive roadblocks that prevent a citizen from properly acquiring and processing the information. Clearly, this circumstance raises many intriguing possibilities for research regarding information processing as it relates to

1 Recent numbers have Fox News Channel winning the cable news wars, with CNN a close second. However, depending on when and in what manner one examines the ratings, either network could be in first place with the other in second place. However, the important fact for this study is that both news networks consistently outpace their cable news competitors.

2 Poll conducted by Garin-Hart-Yang Research Group in November 2004.

1 television news. Therefore, the goal of this study is to gain a deeper understanding of information processing as it relates to information received from the televised news media. Although this examination will focus specifically on Fox News Channel and CNN, the conclusions derived from this study should have implications for individuals who perceive CBS, ABC, NBC, PBS, MSNBC, or any other television news network to be ideologically biased. I will be focusing on Fox News Channel and CNN specifically because survey research has demonstrated that a larger percentage of individuals perceive these two networks to be the most ideologically biased. However, because the nature and effects of perceptions of ideological bias in television news are most likely are similar elsewhere, the information gleaned from the study of these two networks should apply to other news networks as well. Academic Studies of Media Bias Although a number of previous scholars have failed to uncover any pattern of systematic ideological bias within the content of the news, several other types of bias have been shown to pervade the news. In early investigations into this topic, both Weaver (1972) and Hofstetter (1976) found that any bias that existed was driven by independent news standards regarding what is, and what is not, newsworthy. Ultimately, the choice to air (or not to air) a story is driven by editorial consideration of these criteria, rather than being driven by the desire of reporters and/or editors to advance a particular ideological agenda (Graber 1980). Perhaps the most prevalent type of media bias detected by scholars is negativity bias, wherein reporters focus mainly on news events that feature conflict, attack, and oppositional statements, rather than reporting the day’s “good news” (Leighley, 2004). This bias has been demonstrated in numerous examinations of political campaigns (Hallin 1992; Center for Media and Public Affairs 2000; Project for Excellence in Journalism 2000). Robinson and Clancy (1985) demonstrated that this negative tone continues once candidates actually take office, and concluded that the propensity of television news to focus on bad news often results in bias against incumbents (Brody 1991; Grossman and Kumar, 1981). This type of bias has been found to be, for the most part, evenly distributed across partisan lines (Graber 1980; Robinson and Sheehan 1983),

2 and is typically attributed to the journalistic practice of holding incumbents to a higher level of scrutiny because incumbents are actually in a position to directly influence public policy decisions. This overwhelming level of media negativity has played a major role in the emergence of what scholars have referred to as hostile media bias. This was initially identified by Vallone, Ross, and Lepper (1985), who found that partisan viewers tend to perceive media coverage of relevant events as being more supportive of the opposing political position, even though these viewers often could not cite specific instances as to why they believed the news media was biased against their position. The most widely accepted explanation offered by the authors as to why this occurs is that partisanship is preventing individuals from performing an objective analysis of the news content being presented to them (see Peffley, Glass, and Avery 2001). Although confirmed by many (Dalton, Beck, and Huckfeldt 1998; Peffley et al., 2001), the underlying theory has been challenged by those who refute that people perceive the media as hostile toward their position (Giner-Sorolla and Chaiken 1994; Gunther and Christen 2002; Gussin and Baum, 2004). Other scholars argue that a bias toward ease of storytelling is the most prevalent type of bias in television news. In an examination of the types of stories that typically made the nightly news broadcast, Iyengar (1987; 1991) argued that stories are presented episodically, wherein political and social issues are presented as unrelated, discrete events, because television news is better suited to that type of coverage (see also Bennett 1995, as his idea regarding fragmentation bias is in line with Iyengar’s notion of episodic framing). Bennett (1995) reached a similar conclusion, when he argued that news stories were personalized and dramatized for the purposes of making them both seem more important socially and easier for reporters to cover. As was the case with the previous scholars, however, these scholars do not find ideological bias to be pervasive within the news media.3

3 The vast majority of this research predates the origin of Fox News Channel, and some of these studies even predate the emergence of cable news, and thus CNN, altogether. Therefore, it is quite possible that some of these authors would be amenable to the position that Fox News Channel and CNN are ideologically biased.

3 Recently, attempts have been made that specifically examine whether Fox News Channel and CNN are ideologically biased. A report prepared by the Program on International Policy Attitudes (PIPA 2003) found that Fox News Channel viewers were much more likely to have administration-friendly misperceptions about the Iraq War and the War on Terror than were viewers of other news networks. Also, scholars have found that the tone of Fox News Channel’s and CNN’s coverage of the Iraq War and the War on Terror was overly positive and negative, respectively, which may indicate a conservative slant on Fox News’ part and a liberal slant from CNN (Journalism.org 2005). In addition, Groseclose and Milyo (2005) applied the logic of ADA scores in attempting to analyze ideological news bias. The authors found that the majority of media outlets, including all shows studied on CNN, were liberally biased, and one of only two outlets they found to be conservatively biased was “Special Report with Brit Hume”, a staple of Fox News Channel’s nightly lineup. However, with regard to the PIPA study, it has not been demonstrated that these war-related misperceptions can be directly tied to Fox News Channel reporting any of these falsehoods as fact, which casts doubt on whether these misperceptions are a result of the active advancement of a pro-conservative agenda on the part of those at Fox News. Essentially, this study implies that a causal relationship exists when only a correlational relationship has been demonstrated (i.e. people who are misinformed, especially those misinformed in a pro-Bush direction, self-select as Fox News Channel viewers). In addition, the studies conducted by both www.journalism.org (2005) and Groseclose and Milyo (2005) have been criticized due to their reliance on content analysis, as the inherent subjectivity of this methodological approach exposes the researcher to criticism based upon coding choices, which many proclaim to be arbitrary and possibly may lead to systematic error. Given these problems, as well as the findings of the overwhelming majority of the academic studies reviewed here, I am ultimately led to infer that the academic belief that there is little, if any, ideological bias within television news media remains widespread. Criticisms from the Left and Right The most important implication to take away from the aforementioned academic research is that most of these scholars failed to uncover any significant, systematic

4 evidence that the content of television news was ideologically biased. Assuming that these notable scholars are correct, we are left with the question of what factors could account for the chasm between the high level of ideological bias citizens perceive in television news and what scholars have uncovered about the nature of the content presented in television news. Media critics from both sides of the political spectrum offer several explanations. Liberal allegations of a conservative news media bias tend to focus on two specific issues: corporate ownership of most large news media outlets and a conservative shift in news sourcing patterns. First, liberal critics argue that news media ownership by large corporations has led to the U.S. news media being dominated by a profit driven, inherently conservative, pro-business perspective. As a result, many liberals assert that this has led most news media sources to become less critical of both corporate interests and the conservative economic ideas to which these corporations adhere (Chomsky 1987; see also Parenti 1997). Secondly, liberal critics argue that the news media’s heavy reliance on conservative leaning think tanks for information and experts has essentially managed to push liberal ideas out of television news. This claim is supported by a recent study conducted by Fairness and Accuracy In Reporting (FAIR 2003), which demonstrated that conservative think tanks received 47% of all news source citations, as opposed to the 40% received by non-partisan think tanks and the mere 13% of news source citations received by liberal-leaning think tanks. Conservative allegations of a liberal news media bias typically emerge from research into the ideological leanings of news reporters and editors, both of whom conservatives argue tilt their stories in a liberal direction in order to aggressively promote an overly progressive social agenda (Goldberg 2002). These conservative critics cite statistics to support the notion that reporters and editors are disproportionately liberal. According to the Kaiser Family Foundation (2001), reporters are almost five times more likely to be liberal than conservative and are substantially more likely to vote for a Democratic presidential candidate than for a Republican presidential candidate. Conservative critics assert that the statistics regarding the political ideology of reporters and editors are important because there is a real risk that the coverage of events we

5 receive from these journalists is colored by their individual biases and political leanings (Patterson 1994). These arguments have their strengths and weaknesses. The ownership perspective is plausible given the relationship between both managerial figures (Roger Ailes) and on-air personalities (Tony Snow) at the Fox News Channel and the Bush administration.4 However, conservatives would counter that CNN was owned for years by outspoken liberal business entrepreneur Ted Turner, who has never been an advocate of a conservative agenda. Also, the statistics regarding story sourcing and the political leanings of journalists and editors do appear to be sound, yet, the mere fact that a journalist happens to be a liberal does not necessarily mean that he or she cannot cover a news story in an ideologically neutral fashion. The veracity of these claims must also be tempered by the fact that most are made by partisan activist groups, who have a vested interest in advocating on behalf of a particular political position. It is also important to keep in mind that, given the low level of political knowledge that permeates American society (Converse 1964), it is doubtful that much of the American public is aware of many of the characteristics of these major news outlets beyond simply knowing that reporters and news editors tend to be liberal. Finally, perhaps the most important counter to these claims by activists is what was pointed out earlier: the news media may be corporate owned, and many reporters and editors may indeed tend to be liberal leaning Democrats, but systematic research does not find that this matters for the actual content of news stories. What’s Missing For the most part, there have been two different approaches utilized to study the existence of media bias. First, many notable scholars have taken a content analysis style approach to studying this topic. Although, as previously mentioned, these scholars have uncovered the existence of a few types of media bias, most have consistently failed to find any evidence of the existence of ideological bias in the news media (Weaver 1972;

4 Ailes, a former campaign advisor for George H.W. Bush, offered President George W. Bush political advice in the aftermath of the September 11th terror attacks (Woodward 2002). This move was considered taboo by many media critics, given Ailes’ position as chairman of Fox News Channel at the time this advice was given. In addition, former Fox News weekend anchor Tony Snow was recently chosen to be press secretary and advisor to the Bush administration. Many media critics also saw this as an embarrassment to the network.

6 Hofstetter 1976; Graber 1980; Robinson and Clancy 1985; Robinson and Sheehan 1983). A second perspective, whose advocates largely consist of ideologues and activists, presents more qualitative arguments as to why the news media are either liberally or conservatively biased. This group references factors such as the overwhelming liberalism of reporters and editors or the overtly conservative nature of big-business media ownership groups as the main reasons why these media outlets are either liberally or conservatively biased. In this study I do not argue that these research approaches are incorrect. In fact, I believe these approaches are meritorious efforts to address the questions they seek to answer. However, despite their merit, I do believe that they fall short in one very important aspect. These studies focus on demonstrating whether ideological bias in the news media exists, rather than attempting to explain its implications once it is perceived to exist. It is quite possible that ideological bias in the news media is a case in which perception is more important than reality. If Fox News Channel and CNN are ideologically biased, but no one knows or cares, then perhaps we should not care either. On the other hand, if everyone perceives these two networks to be ideologically biased, then, regardless of the reality, those perceptions may be influential, and thus worthy of scholarly inquiry. Specifically, I hope to determine whether individuals are relying on network attribution as a heuristic device to determine the ideological content of a news message. It could be the case that individuals recognize that news content is from Fox News Channel and CNN and identify that news content as conservatively and liberally biased, respectively. If this type of judgment is taking place, it is also quite possible that not only do these cues influence perceptions of ideological bias, but that reliance on these cues may also influence individuals’ resulting ability to acquire and retain the news content presented by these news sources. The Messenger Overwhelming the Message Thus far, I have argued that it is doubtful that individuals who perceive ideological bias in the news media are typically responding to any ideologically biased content within the news broadcast, based upon the number of credible scholarly studies that have found that this type of news media bias is not prevalent. Also, the specific

7 claims regarding ideological bias in the media made by political activists and ideologically driven “watchdog” groups have been shown to be suspect at best. This returns us to the original question of interest for this study: Why do individuals assert that Fox News Channel and CNN are conservatively and liberally biased, respectively? I argue that individuals are relying on the network cue as a heuristic to provide a signal as to whether the news content presented is either conservatively or liberally biased. In short, once it became part of the conventional wisdom that Fox News Channel and CNN are biased, citizens’ perceptions of those networks followed the conventional wisdom— possibly irrespective of whether these networks’ shows actually featured any ideologically biased content. The theoretical underpinning for this study is derived from the findings of research done on the benefits and drawbacks of the use of heuristics by Kuklinski and Hurley (1994; see also Asch 1952, McGuire 1969, and Zaller 1992). Of particular interest for this study is the phenomenon which is perhaps best characterized as the “messenger overwhelming the message”, wherein an ideological component (Joslyn and Ceccoli 1996; Entman 1989; Dalton, Beck, and Huckfeldt 1998) and source cues (Carmines and Kuklinski 1990; Mondak 1993a, 1993b) are combined in the evaluation of political messages. In their experiment, Kuklinski and Hurley (1994) analyzed differences in levels of support within the white and African-American communities for the statement that “African-Americans must stop making excuses and rely much more on themselves to get ahead in society.” In this experiment, the message respondents received remained constant, while the messenger who delivered the message was manipulated. This statement was attributed to George Bush, Jesse Jackson, Ted Kennedy, Clarence Thomas, or no specific individual. If the content of the message is all that matters, then to whom the message is attributed should not have a significant influence on the interpretation of the message. However, if the messenger is having an effect, then there should be significant differences in the interpretation of the message based on the source to which that message is attributed. The results of this study demonstrated that the interpretation of messages is often influenced by the recipient’s evaluation of the reputation of the individual who is providing the message. Even though the content of the message did not change, African-

8 American respondents were much more accepting of this message when it came from the individual to whom they were closest politically (Jackson) than when it came from the individual with whom they most disagreed politically (Bush). In addition, the race of the messenger was also found to have a profound impact on the interpretation of the message, as African-American respondents demonstrated substantially lower levels of anger in reaction to this statement when attributed to Jackson, Kennedy, or Thomas than when it was attributed to Bush. This research is supported by numerous subsequent studies that have reached similar conclusions (Druckman 2001; 2001b; 2001c; Gunther and Schmidt 2004). By taking a novel approach to the study of cue taking, these authors were able to demonstrate that, with regard to political information, the messenger matters. I believe that this same approach can be successfully utilized to determine why individuals perceive ideological bias in the television news media. As previously mentioned, scholars have used content analysis to address the issue of ideological bias in the news media, typically concluding that this type of bias is not pervasive. However, the question of why individuals still perceive ideological bias in the news media despite these findings remains unanswered. Perhaps, much like the statement regarding African-American self- reliance, the substantive content of television news broadcasts is not the major issue; perhaps of more importance is the viewer’s evaluation of the ideological leanings of the news network that is delivering that content. I believe that utilizing the findings regarding the “messenger overwhelming the message” will allow me to obtain leverage on this critical question. In addition, this approach will provide insight into how these perceptions of ideological bias impact information acquisition and retention. If, as I suspect, the network label is sending a signal that influences perceptions of ideological bias, it is likely that this same signal is influencing the manner in which respondents process that information. Building mainly upon the approach of Kuklinski and Hurley (1994), the central claim of this study is that people evaluate newscasts as biased as a reaction to the network cue as opposed to any actual ideologically biased content within the newscasts, and that this evaluation influences the extent to which individuals acquire and retain information. To determine whether the “messenger” (network cue) is overwhelming the “message”

9 (substantive content of the news stories) with regard to news content presented by Fox News Channel and CNN, I employed a multistage research design. First, I videotaped several hours of news footage from CNN and Fox News Channel over a three-day period in January 2004. Then, I reviewed the videotaped footage from each network and selected five stories from each network for this analysis. Finally, I devised and conducted an experiment in which I re-created the original CNN and Fox News Channel broadcasts, while manipulating the network attribution for each set of stories. Holding news content constant while manipulating the network attribution will allow me to investigate whether the CNN and Fox News Channel labels are actually sending ideological cues to the viewer and thus biasing both their perception and acquisition of the content presented. An experimental research design is, I believe, both a creative and appropriate approach to answering these questions. However, simply being a creative approach to answering a specific problem does not mean that a study should be conducted. Even the most inventive study need not be conducted if that study fails to make an important substantive contribution to the field’s current state of knowledge. I argue that this study provides new insights on many theoretically important questions. First, this approach will provide insight into questions about ideological media bias. Specifically, it will allow me to address whether individuals really believe that Fox News Channel and/or CNN are ideologically biased, and, if so, whether this perception is driven by the news content these networks present, the ideology of the viewer, the alleged ideology of the network, or some other yet undetermined factor. In addition, this study will provide insight into how these perceptions of ideological bias influence an individual’s ability to “learn” from the news. This is particularly important given the news media’s recognized role as the political informant for the American public. Does the perception of ideological bias in the news media inhibit or enhance learning, and why? This examination will allow me to gain leverage on these questions and will, I believe, make a valuable contribution to this field of study. Chapter Outlines This study will proceed in the following manner. First, the experimental design employed in this examination will be described in detail in chapter two. This explanation

10 will, most importantly, allow me to demonstrate why an experimental approach is the appropriate method to employ to answer this research question. In this chapter I will illustrate the specific details of how this study was conducted - - the process by which the news stories used were chosen, the manner in which the news broadcast replications were created, how the sample used for this examination was chosen, and, ultimately, how the experiment was implemented and data collected. Also, I will present detailed manipulation checks that will demonstrate that the experimental manipulations achieved their desired effect. Finally, in this chapter I will address (and hopefully alleviate) any potential concerns regarding this experimental design. Chapter three begins the empirical analysis with an examination of what drives viewer perceptions of ideological bias within the news media. Previous scholars who have examined this controversial topic have found little or no systematic evidence of ideological bias in the media. However, survey research has demonstrated that citizens do in fact perceive ideological bias in the media, specifically with regard to CNN and Fox News Channel, which allegedly represent the liberal and conservative political viewpoint, respectively. In this chapter I will argue that attaching the CNN and Fox News Channel labels to news stories is essentially the equivalent of sending an ideological cue to the viewer regarding the content of the story. I anticipate that knowing that a news story is being presented by either Fox News Channel or CNN will cause viewers to perceive the stories as being conservatively or liberally biased, respectively, regardless of the actual news content presented. By manipulating the network attribution, I am able to demonstrate that this is indeed the case, as the CNN and Fox News Channel labels do indeed function as ideological signals to the viewer, with these signals being most pronounced among ideologues whose views are supposedly at odds with those attributed to the network. Chapter four serves as the second empirical chapter. In this chapter I investigate what, if any, influence perceptions of ideological bias have on viewers’ ability to acquire and recall information presented to them by Fox News Channel and CNN. It is possible that being exposed to a consonant media source could lead individuals to pay more attention to the news content being presented to them, whereas they would be more inclined to dismiss the news content being presented to them by a dissonant media

11 source. Specifically, conservatives and liberals would better acquire and recall news content presented by to them Fox News Channel and CNN, respectively. In contrast, however, it is also possible that individuals may pay extra attention to a dissonant media source precisely because it is dissonant, and not pay as much attention to a consonant network, because they just assume that they will agree with the information presented. If this is the case, conservatives and liberals would better acquire and recall news content presented to them by CNN and Fox News Channel, respectively. Ultimately, I am able to empirically demonstrate that individuals are better able to acquire and recall news content presented to them by a dissonant media source than news content presented to them by a consonant media source. Chapter five concludes the empirical analysis. In this chapter I investigate why, even when presented with exactly the same news content, individuals are better able to acquire and retain factual information when that news content is attributed to a news network with which they disagree politically. Believing that a source of news content is ideologically dissonant appears to activate an underlying cognitive tendency among individuals that inspires them to pay more attention and, ultimately, to acquire more information. I argue that this occurs because the respondent formulates counterarguments in response to the information with which he or she is presented in the news broadcast. Respondents believe that the dissonant network they are watching is hostile toward their political position, and, as a result, they remember more of the factual information presented in order to be able to demonstrate why the position taken by this dissonant news source is incorrect. In this instance, I anticipate that conservatives will formulate counterarguments more often in response to news content presented by CNN and that liberals will formulate counterarguments more often in response to news content presented by Fox News Channel. My empirical findings lend strong support to these theoretical expectations. In the final chapter I will address the implications of these findings within the context of the theory. This chapter will begin with a brief summary of the empirical results presented in this study. That will be followed by a discussion of the various theoretical implications of these findings. I will then provide a discussion of what these results tell us about the ability of the news media to serve both as a linkage institution and

12 as the political informant of the American public, as well as what these results tell us about the ability of the American citizenry to process political information presented on television news broadcasts. Finally, I will conclude this chapter by providing recommendations for future avenues of research that would enable us to gain greater insight into this often discussed, yet not well understood, topic in contemporary political science.

CHAPTER TWO

RESEARCH METHODOLOGY

With a few noteworthy exceptions (Gosnell 1927; Eldersveld 1956), early political scientists shied away from conducting experimental research. Many researchers voiced ethical concerns regarding experiments, such as maintaining participant privacy, the right of the participant to withdraw from the experiment, and the morality of denying one group of experimental participant’s access to the fruits of a potentially beneficial social program (Rivlan and Tempane 1975). Experimental research was often eschewed because it can be much costlier than non-experimental research on the same subject (Burtless 1995), particularly because it is often necessary to provide individuals with financial incentives to participate. Researchers also worried about the representativeness of the sample under study, as many thought that respondent-level differences between those who decided to participate, and those who decided not to participate, would ultimately result in a nonrandom sample of the population under study (Burtless and Orr 1986). In addition, the small size of most experimental samples led many researchers to question the generalizability of experimental results, especially with regard to important decisions, such as the implementation of national public policy programs (Garfinkel et al. 1992). As a result of these, and other, factors, for many years the use of experiments in conducting research on political behavior has been limited. In the past generation, however, researchers have shifted their focus toward the opportunities that experiments provide rather than toward the problems this research approach poses. As a result, experiments have become much more prominent in political science, and their use has dramatically affected this field of study. Experiments have become beneficial tools for social scientists due in large part to the amount of control this type of design affords the researcher. First, an experimental design permits the

14 researcher to create the necessary test condition rather than having to hope that the condition will eventually occur in the real world (Iyengar, Peters, and Kinder 1982). By creating this condition, the researcher is able to hold constant all extraneous factors and isolate the specific effect of the independent variable(s) on the dependent variable(s) under investigation. Also, in many instances experimental research designs provide analytical leverage on many problems that observational data either cannot or does not adequately address, such as the causes of voter turnout decline (Gerber and Green 2000), the causes of political decision making (Lodge, McGraw and Stroh 1989; Lodge, Steenbergen, and Brau 1995; Lau and Redlawsk 1997), and the relationship between the television news media and public opinion (Iyengar and Kinder 1987). Finally, experiments permit the investigator to randomly assign participants, which helps ensure that selection bias does not taint the experimental design and thus bring the researcher’s findings into question (Burtless 1995). As a result, by both creating an experimental manipulation that holds constant all extraneous factors and by randomly assigning research participants, researchers using an experimental design can ensure that their findings are both theoretically significant and not attributable to some unforeseen external factor (Iyengar and Kinder 1987). An experimental design is employed here precisely because it allows the researcher a substantial amount of control over the variables of interest. To reiterate, the theoretical expectation for this study is that citizens are identifying news content presented by Fox News Channel and CNN as ideologically biased due to the network attribution rather than any underlying ideological bias present within the content of the news stories, and this perception significantly influences the manner in which individuals process information presented on television news. Therefore, in this instance an experimental approach is extremely beneficial as it would be impossible to separate the CNN and Fox News Channel labels from their actual content because, in the real world, there is a perfect correlation between the airing of a substantive news story by CNN (or Fox News Channel) and the presence of the CNN (or Fox News Channel) cue. Hence, investigating whether these network attributions are indeed sending ideological cues to the viewer would be impossible because there would be no means to determine if viewers’ perceptions of ideological bias were driven by the substantive content of the

15 news stories, the ideological signals transmitted via the network cue, or a combination of the two. However, an experimental setting will allow me to separate network attribution from substantive news content presented and thus investigate my theoretical expectations. Overview of the Experiment To determine whether the network cue is overwhelming the substantive content of the news stories with regard to news content presented by Fox News Channel and CNN, I employed a multistage experimental design. The experiment involved creating replications of original CNN and Fox News Channel broadcasts, wherein I manipulated the network attribution for each set of stories. To conduct this experiment, I first videotaped several hours of news footage from CNN and Fox News Channel over a three-day period in January 2004. I then reviewed the videotaped footage and selected five stories from each network for this analysis. The final step involved recruiting individuals to serve as experimental subjects, conducting the experiment, and collecting the data. I believe this design is an effective approach to examining this problem because manipulating the network attribution will allow me to separate substantive news content presented by the networks from network attribution and, as a result, investigate whether the CNN and Fox News Channel labels are actually sending ideological cues to the viewer and thus biasing his or her perception of the news content presented. Story Selection The first stage of this process involved deciding which stories from Fox News Channel and CNN to present to my research subjects. I videotaped several hours of news footage from CNN and Fox News over a three-day period in January 2004. Then, I reviewed the videotaped footage and selected five stories from each network for this analysis. The five CNN stories that were chosen were about the war in Iraq, nuclear inspections in North Korea, a female suicide bomber in Israel, a newly proposed “support of marriage” initiative, and Saddam Hussein. The five Fox News Channel stories that were chosen were about the war in Iraq, nuclear inspections in Libya, mad cow disease, illegal aliens, and Saddam Hussein.5

5 The text of the original CNN and Fox News Channel stories used in this examination can be found in the Appendix of this chapter.

16 A number of factors were considered when deciding which stories from the videotaped broadcasts to select for this analysis. First, I eliminated stories through two mechanisms. Stories that featured highly recognizable “talking head pundits”, such as Fox News Channel’s Sean Hannity, Bill O’Reilly, or Alan Colmes and CNN’s James Carville, or Paul Begala, were not used.6 These pundits present “commentary” or “news analysis,” whereas my concern here is whether the actual hard news stories aired by CNN and Fox News Channel convey ideological cues. Second, stories that contained a substantial amount of back and forth banter between the reporter and anchor were not chosen because of the logistical difficulties of replicating the anchor/reporter interaction. The final selection of stories came from those left over after the initial elimination process. This was admittedly a subjective decision, as there were stories that were not chosen that could have easily been substituted for the ones that were chosen. Those ultimately chosen were ones that closely mimicked the coverage one would see on a broadcast during the period in which this study was conducted (stories that cover domestic politics, foreign affairs, health concerns, human interest, etc.).7 News Replications The second stage of this experimental design required the creation of believable replications of these original news broadcasts. Creating believable replications was essential in order to maximize the validity of this study. To achieve this end, the replications were created at a professional television studio (WSST-TV 55 in Cordele, GA), relying on the assistance of Phil Streetman, an accomplished professional television news broadcaster with over twenty-three years of television news broadcast experience. In addition, using a professional television studio gave the videos the appearance of an

6 Although Paul Begala and James Carville are no longer employed by CNN, they were prominent on-air personalities at the network during the time this study was conducted.

7 The State of the News Media 2004 study, conducted by Journalism.org, confirmed that foreign policy stories were a major part of cable news broadcasts when this study was conducted, as this type of story constituted 24% and 21% of the broadcast time on CNN and Fox News Channel, respectively. However, this is not always the case, which raises questions regarding whether these results extend to all periods and forms of news coverage. Within the confines of my data, I can determine whether respondents reacted differently to domestic and foreign news stories. They did not, as the difference of means between foreign and domestic stories originating from CNN and Fox News was only .16 and .15, respectively, and not statistically significant. Therefore, it does not appear that the mix of domestic/foreign stories influences the nature of results one way or the other. However, additional attention is warranted regarding the possibility that perceptions of ideological bias differ for domestic vs. foreign news stories.

17 authentic television news broadcast, and thus lessened the probability that respondents would be able to determine that what they were watching was not an original news broadcast. To create the most authentic replications possible, the anchor read transcripts from the original Fox News Channel and CNN stories in order to duplicate the exact language used in the stories. The experimental manipulations were achieved by the broadcaster vocally identifying with which news organization he was affiliated. At the outset of the broadcast, the anchor would “sign on” as representing the specific network, stating “this is Phil Streetman reporting for CNN” or “this is Phil Streetman reporting for Fox News Channel.” The anchor would reinforce his network affiliation during the news broadcast by using phrases that convey attribution but do not color the ideological tone of the broadcast, such as “CNN has learned…” or “sources have told Fox News that…”. Finally, the anchor would sign off using the tagline for each network. When identifying with CNN, the anchor would state “this has been Phil Streetman for CNN, the most trusted name in news.” When identifying with Fox News Channel, the anchor would state “this has been Phil Streetman for Fox News, where you always get fair and balanced coverage.” The broadcaster recorded each set of stories from CNN three times: once identifying himself as being affiliated with CNN, once identifying himself as being affiliated with Fox News Channel, and once giving no network affiliation. Likewise, the broadcaster recorded the set of stories from Fox News Channel three times: once identifying himself as being affiliated with CNN, once identifying himself as being affiliated with Fox News Channel, and once giving no network affiliation. Each replication set had a running time of roughly ten minutes. It is important to note that any artificiality that remains in the replications should constitute a conservative bias for hypothesis testing. Although the news stories were produced in a professional news studio with the help of a professional anchor, the studio lacked the characteristic trappings of the major cable news networks. Also, network affiliations were conveyed via verbal reference to the networks, but these references were not reinforced with the network news logos. In short, Fox News Channel and CNN broadcasts make it even clearer what network the viewer is watching, and the fact that my

18 replications could only approximate a portion of this cueing effect means that any ideological signals transmitted in my tests almost certainly were weaker than those conveyed by actual Fox News Channel and CNN broadcasts. Recruiting Experimental Subjects A sample of 246 undergraduates enrolled in political science courses at Florida State University participated in this experiment in the spring of 2004. Experimental participants for this study were recruited from various introductory and upper-level undergraduate political science courses. When recruited, participants were simply told that they would be participating in a news media focus group. Subjects were offered extra credit in the course in which they were enrolled as an incentive to take part in the study.8 Those who volunteered initially had the ability to opt out of the study (i.e. to not show up for their assigned time or refuse to complete their participation). However, those individuals who volunteered to participate but, for whatever reason, either failed to appear or did not fully participate in the experiment, did not receive extra credit in their course. It is reasonable to ask what implications arise from using a student sample for this study. There is a risk that a younger student sample would possess lower levels of political knowledge and interest and thus not be representative of a larger population (Hovland 1959). Therefore, it would constitute a more conservative test of my hypothesis than a random non-student sample. Although this is a design concern, I have taken care to mitigate its tangible effect on the results. Specifically, this sample was selected from various undergraduate political science courses (with many of them being political science majors enrolled in upper level courses). As a result, we would expect their levels of political knowledge and interest to be higher due to their interest in the subject as well as their information exposure levels. Therefore, although not the optimum random non-student sample, these respondents should not be plagued as severely by the “college sophomore” problems typically associated with the use of students as research subjects.

8 Professors teaching the courses from which the experimental subjects were recruited had full discretion over the amount and form of extra credit their students received for participating in the experiment (i.e. students were rewarded with different levels of extra credit depending on who their professor was). However, I have no reason to believe either that these students were aware of the fact that they received different types of extra credit or that this difference in any way affected the responses given by participants.

19 Implementation and Data Collection As Table 2.1 illustrates, the core of the experimental design is a two by three factorial, and the manipulated variables are the network that actually produced the story (CNN or Fox News Channel) and the network to which the stories are attributed (CNN, Fox News Channel, or no attribution).9 One videotape consisted of five CNN stories that were attributed to CNN, another consisted of five CNN stories that were attributed to Fox News Channel, and a third contained five CNN stories that were given no network attribution. Three other videotapes provided analogous manipulations for Fox News Channel stories. Participants were randomly assigned to one of the aforementioned experimental conditions. This was achieved by matching the name of a randomly selected volunteer to a randomly chosen experimental treatment. Table 2.2 illustrates that, in this instance, randomization was very successful. The randomization process resulted in the eight experimental conditions being roughly equal with regard to the gender, party identification, and political ideology of respondents. This randomization process also ensured that the size of each experimental cell was roughly the same. After viewing the videotaped news stories, participants completed a survey instrument. This survey instrument contained questions that achieved a number of goals. The most important aspect of the survey was that it provided a measurement of several potential dependent variables. It also provided a substantial amount of potentially useful background information by administering a battery of questions regarding socio- demographic and personality characteristics. The survey instrument was not administered until after the subjects had viewed the entire video, both so respondents could devote their full attention to the videotapes and to avoid signaling to the respondents what they should specifically be looking for in the videotapes in order to answer the questions. The survey served two valuable purposes in addition to providing measures of the study’s variables of interest. First, the survey helped mask the actual purpose of the

9 The original CNN and Fox News Channel broadcasts were also shown to some participants, and thus there actually were eight cells in the design. As previously noted, the originals were used as part of a manipulation check, and data from these cells were eventually incorporated into the core 2 X 3 design.

20 study. If you recall, participants were told that they were participating in a news media focus group. By including related questions I was able to hide the actual purpose of the study within a broad survey instrument. Only after the completion of the experiments were respondents alerted to the more specific goals of the study. Also, as a check to make sure respondents were not keying in to what the study was actually about, I concluded the survey by asking them to make a guess as to the specific purpose of the project. There was no consensus answer to this question; responses ranged from the cover story they were told (this was the most cited response) to the admission of no idea. Most importantly there was no systematic response indicating that participants were aware of the actual intent of the experiment. Only ten of the 246 respondents correctly guessed some variant of “media bias study.” Because this response indicated the possibility that these respondents were “cued in” to the specific aims of the study, these ten respondents were removed from the analysis. After all respondents completed their participation in the experiment, they were informed on the true purpose of the study.10 Manipulation Checks To draw valid conclusions from this study regarding the hypotheses, it was essential that the participants were responsive to the experimental manipulations implemented in this study. The first question is whether participants noted and accepted the experiment’s network attributions. After all, it would be meaningless to attribute news stories to CNN, Fox News Channel, or any other network were it the case that viewers tune out when such network affiliations are mentioned. To test this, I asked participants what network they had viewed. What we should find is that participants who viewed stories attributed to Fox News Channel and CNN accurately reported those attributions. As Table 2.3 illustrates, this indeed did occur, as 92.4% of the experimental subjects retained the experimental manipulation. Beyond serving as a manipulation check, this finding has two important substantive implications. First, the high rate of success for the manipulation indicates that it was not difficult to convince subjects that CNN content came from Fox News and vice versa. Therefore, any

10 This study was conducted in accordance to the guidelines set forth by the Office of Research at The Florida State University. For more information on these guideline visit the Office of Research website located at www.research.fsu.edu/humansubjects/index.html.

21 perceptions of ideological bias can be directly attributed to the ideological signal the network sends to its viewers and not to any underlying bias within the content of the story. Secondly, those who did not retain the experimental manipulation were dropped from the analysis, because people who do not grasp or do not care what network they are watching should essentially be immune to any cueing effects from the network attribution. Because the major claim of this study is that viewers perceive CNN and Fox News Channel as the most liberal and conservative television news organizations, respectively, it is important that my sample views these two networks in that way. Table 2.4 indicates that this is indeed the case. Keep in mind that respondents were asked to rank these networks after they were asked about their perceptions of the stories in order to ensure that their thoughts on the networks were not influencing their opinions regarding the stories. Also, in this analysis the respondents who received an attribution were separated from those who did not in order to control for the possibility that people may have been judging the networks based on the content of the stories they just viewed (i.e., a respondent believes news stories are liberally biased, knows stories are from CNN, and therefore ranks CNN as being more liberal than he/she initially would have). Roughly one quarter of respondents viewed news stories without network attributions, and thus their perceptions of the ideological tone of the networks cannot be influenced by the experiment itself. It turns out that this distinction makes little statistical or substantive difference in the analysis. The reference point for these groups is a score of five, which represents perfect ideological neutrality for the network. In both portions of the analysis, participants perceived CNN as the most liberal and Fox News Channel as the most conservative of the major news organizations. These networks are also the only ones whose evaluations are significantly different from the point of ideological neutrality. This is substantively important because it justifies my choice of networks for this analysis. As a final manipulation check, I compared the original CNN and Fox News Channel broadcasts with the replications in order to ensure that there was not any discernible difference between the two that may be influencing any potential evaluation of ideological bias.

22 The original versions of these stories were compared with the replications on six different criteria: anchor appearance, anchor tone of voice, anchor level of knowledge, anchor trustworthiness, anchor objectivity, and story ideology.11 Table 2.5 illustrates that the only significant differences that emerge involve anchor appearance and tone of voice. There is no significantly discernible difference in anchor level of knowledge, trust, objectivity, and story ideology, between the originals and the replications. This is important for two reasons. First, it indicates that the replications were successful in capturing the ideological tone of the substantive content of the news presented in the original broadcasts. Second, the fact that no difference exists in most categories (specifically story ideology) allows me to pool the originals with the replications that were attributed to the proper network and analyze them within the two by three factorial framework. Addressing Validity Concerns A concerted effort was made to enhance the external validity, and thus the generalizability, of this study. Two steps were taken to guard against “demand characteristics”, or cues that alert participants to what is expected of them (Orne 1962). First, I took great care to hide the true purpose of my experiment. I began each experimental session by telling participants a believable, yet false, cover story regarding the true purpose of the experiment. In this case, participants were told that they were participating in a focus group about television news media. In addition, questions which addressed the issues of interest for this examination were embedded within a lengthy survey instrument. By providing this plausible cover story and reinforcing it with questions that actually asked respondents a variety of media related questions, I hoped to prevent them from digging deeper to try to uncover the actual agenda of this study. Second, I took steps to try to minimize the “artificiality” that is common with experimental research. Creating the replications by using a professional news anchor in a professional television studio gave the videos the appearance of a real news broadcast and thus lessened the probability that respondents would be able to determine that the

11 “Anchor appearance” served as a measure of how telegenic the anchor was to the respondent. “Anchor tone of voice” served as a measure of how well the anchor vocally delivered the news. “Anchor level of knowledge” measured how intelligent the respondent thought the anchor was. “Anchor trustworthiness” measured whether the respondent trusted the anchor, and “anchor objectivity” measured whether the respondent thought the anchor was delivering the news in an ideologically neutral manner.

23 video they watched was not a recording of an original news broadcast. However, again, it is important to note that any artificiality that remains in the replications should constitute a conservative bias for hypothesis testing. Although the replications were produced in a professional news studio, the studio lacked the characteristic trappings of the major cable news networks. Also, network affiliations were conveyed via verbal reference to the networks, but these references were not reinforced with the network news logos. In short, Fox News Channel and CNN broadcasts make it even clearer what network the viewer is watching, and the fact that my replications could only approximate a portion of this cueing effect means that any ideological signals transmitted in my tests almost certainly were weaker than those conveyed by actual Fox News Channel and CNN broadcasts. Conclusion Great care has been taken with regard to the design and implementation of this experiment. Despite this level of care, it is important to keep in mind that the risks associated with generalizing experimental results to large populations can never be completely eliminated. However, by taking steps to reduce the impact of demand characteristics on my experimental subjects, and by limiting the amount of artificiality that exists in the news broadcasts replications, I believe I have minimized as much as possible the methodological problems associated with generalizing from experimental results.

24 Table 2.1 Story Source by Network Attribution12

Source Attribution Manipulation

CNN Original Version 1

FNC Original Version 2

CNN No Attribution 3

CNN CNN 4

CNN FNC 5

FNC No Attribution 6

FNC FNC 7

FNC CNN 8 ______

12 In the 3X2 factorial, the original version of CNN was combined with the CNN replication that was attributed to CNN. Likewise, the original version of Fox News Channel was combined with the Fox News Channel replication that was attributed to Fox News Channel.

25 Table 2.2 Experimental Cell Demographics

CNN* FNC* CNN/FNCA CNN/NA FNC/CNNA FNC/NA

N 51 52 35 38 37 35

Caucasian 37 37 26 17 24 22

Hispanic 9 7 2 9 3 4

African-American 4 6 4 8 7 7

Liberal 19 19 12 16 13 12

Conservative 17 18 13 13 14 14

Moderate 15 15 10 9 10 9

Republican 24 24 15 14 11 15

Democrat 20 21 13 14 15 12

Independent 7 7 7 10 11 8

Male 28 27 17 19 18 19

Female 23 25 18 19 19 16

*The demographic statistics for CNN and FNC include both the originals and the replications.

26 Table 2.3 Experimental Condition by Perception and Retention

Experimental Condition13 Perceived Attribution % Retained14

CNN Story/CNN Attribution Perceived CNN 95.9

FNC Story/FNC Attribution Perceived FNC 95.9

CNN Story/No Attribution No Perceived Attribution 89.1

CNN Story/FNC Attribution Perceived FNC 90.9

FNC Story/No Attribution No Perceived Attribution 88.2

FNC Story/CNN Attribution Perceived CNN 91.6 ______

Perceived Correct 92.4

13 The CNN Content/CNN Attribution and FNC Content/FNC Attribution categories contained both the originals and replications attributed to the respective networks. 14 The 18 respondents who did not retain the manipulation were dropped from the analysis.

27 Table 2.4 Perceived Ideological Bias of Major Television News Networks

Network No Attribution Repsondents All Respondents

CNN 3.87* 4.08**

FNC 6.82** 6.77***

ABC 4.99 5.06

CBS 5.23 5.18

MSNBC 5.27 5.10

NBC 5.05 5.01

N 58 218 ______Note: Superscripts indicate means that are significantly different from 5.00 (perfect ideological neutrality); *p<.05 **p<.01 ***p<.001.

28 Table 2.5 Perceptions of Attributed Stories, Originals Versus Replications

Characteristic of Broadcast Original Replication

CNN N=47

Anchor Appearance 7.67 5.88**

Anchor Voice 7.71 6.04**

Anchor Knowledge 7.43 6.64

Anchor Trust 6.48 6.52

Anchor Objectivity 6.14 5.96

Story Ideology 20.86 20.12

FNC N=47

Anchor Appearance 7.00 6.15*

Anchor Voice 6.88 6.27

Anchor Knowledge 6.96 6.46

Anchor Trust 5.92 6.23

Anchor Objectivity 6.08 5.58

Story Ideology 29.96 30.27 ______Note: *p<.05 **p<.01.

CHAPTER THREE

THE MESSENGER OVERWHELMING THE MESSAGE: IDEOLOGICAL CUES AND PERCEPTIONS OF BIAS IN TELEVISION NEWS

Surveys have demonstrated that news coverage provided by CNN and Fox News Channel is perceived by many Americans as being liberally and conservatively biased, respectively.15 Because a number of Americans rely on the television news media in general, and these networks specifically, as their primary source of political information (Bower 1985), it is important to evaluate whether this perceived ideological bias influences how people process television news. It may be the case that attaching the “CNN” and “Fox News Channel” labels to news stories is virtually the equivalent of sending ideological signals to the viewer. Knowing that a story is from one of those networks may send signals to the viewer regarding the ideological nature of the news content, which could lead the viewer to perceive ideological bias and possibly raise cognitive roadblocks that prevent him or her from properly acquiring and processing the information. Therefore, the goal of this chapter is to gain a deeper understanding of why many citizens perceive ideological bias in television news, despite the fact that this type of bias has typically gone undetected in systematic scholarly inquiry. Heuristic Processing and Television News Interpretation There is certainly disagreement on whether ideological bias exists in the news media and, if it does exist, what form it takes. A survey conducted by the Program on International Policy Attitudes (PIPA) (2003) demonstrated that FNC viewers were more likely to hold misperceptions about the Iraq War and the War on Terror, which has resulted in allegations that FNC advances a conservative agenda. In contrast, Groseclose

15 Although Fox News Channel has recently been on top in the competition for ratings, it is conceivable that either network could be in first place with the other in second place, depending on when one examines the ratings. The important point, however, is that both networks outpace their cable news competitors.

30 and Milyo (2005) assert that nearly all of the news media are liberally biased. Using a technique wherein they calculate scores based on the ADA metric for a number of news media outlets, the authors argue that the news media, with the exception of Fox News Channel’s Special Report with Brit Hume and The Washington Post, advance a liberal agenda. Although these studies are indeed provocative, they have been subjected to scrutiny. For instance, the authors of the PIPA study are unable to demonstrate that misperceptions about the Iraq War and the War on Terror emerged because falsehoods were reported by FNC. Therefore, as is noted in the PIPA report, it is premature to assume causation (and thus ideological bias) based on this study when only a correlation has been demonstrated. It is entirely plausible that characteristics of viewers, not of the news being viewed, drove the observed variance in war-related misperceptions. In addition, the Groseclose and Milyo study has received criticism from those who take issue with their methodology and coding choices.16 Despite the findings of a few recent studies, the vast majority of previous scholarly research provides reason to believe that the actual content of television news broadcasts is not ideologically biased (Graber, 1980; Entman 1989; Robinson and Sheehan, 1983; Robinson and Clancy, 1985; Weaver, 1972; Hofstetter, 1976; Iyengar and Kinder 1987; Iyengar 1991; Bennett and Entman, 2000). However, irrespective of what most of this previous research has demonstrated, a substantial portion of the American public still perceives CNN and FNC as being ideologically biased. Hence, we are left with an intriguing discrepancy. Systematic academic research struggles to find more than a hint of ideological bias in the news, yet the perception that such bias exists is widespread among the mass public. The critical assumption motivating the current research is that this perception of ideological bias can be highly consequential irrespective of whether ideological bias exists in reality. For the viewer who perceives CNN and FNC as advancing liberal and conservative views, respectively, those networks’ labels function as powerful ideological cues. Therefore, simple attribution of a story to CNN or FNC can activate a cognitive heuristic, shaping how given stories are

16 Alterman, Eric (2006). Think Again: Rigging the Numbers. Center for American Progress. http://www.americanprogress.org/issues/2006/01/b1347483.html

31 perceived. Importantly, provided that viewers assume ideological bias to be present, this heuristic process can be activated irrespective of whether the actual content of the news is ideologically biased. In short, perception may trump reality. In social psychology, research on cognitive heuristics has been developed largely with reliance upon dual-process models of information processing. A notable theoretical model in this genre is Petty and Cacioppo’s (1986) Elaboration Likelihood Model, wherein the authors argued that individuals process information via either a central or a peripheral route. Central route processing refers to methods of evaluation wherein the individual carefully considers the information presented when making a decision. By contrast, individuals making decisions via the peripheral route make judgments based on simplistic reactions to stimuli that are, in many cases, only tangentially related to the message at best. One of these peripheral methods involves the use of heuristic processing (Chaiken 1980; 1987; Chaiken, Liberman and Eagly 1989; see also Tversky and Kahneman 1974). According to Sniderman et al. (1991), heuristics are efficient tools for the decision maker to utilize because they require little information while providing answers to complex problems. Essentially, reliance on heuristics allows the individual to evaluate information without having to engage the information in an in- depth manner. Because most individuals deal with matters in their daily lives that are of much greater personal importance to them than analyzing politics, it is highly likely that individuals employ heuristics when trying to understand political news. In this case, viewers appear to rely on two complementary heuristics when interpreting a signal that they attribute to either CNN or FNC. The first of these, the ideological heuristic, speaks to the content of the news. Early social psychology research demonstrated that individual motives, such as ideology, can shape the way individuals perceive the things they see (Hastorf and Cantril 1954). In the field of political science, Joslyn and Ceccoli (1996) demonstrated that the effectiveness of media messages relies on the relationship between viewer ideology and the valence of network news coverage. Entman (1989) and Dalton, Beck, and Huckfeldt (1998) found that perception of news is shaped as much by a person’s political views as by objective content. Scholarly work on hostile media bias, wherein those with entrenched political beliefs tend to believe that the media are hostile toward their political position, also reinforces this idea (Vallone, Ross,

32 and Lepper 1985). As a result, ideology appears to exert a strong, consistent influence on the interpretation of news broadcasts. The second heuristic at work here is source cues. Source cues should be significant because they provide an easy way for an individual to evaluate a message (Carmines and Kuklinski 1990; Mondak 1993a, 1993b). In an early study, McGuire (1969) concluded that people evaluate a candidate and either accept or reject that candidate’s message based solely on a heuristic evaluation. In addition, Zaller (1992) demonstrated that when people process information they use the reputation of the message provider as a cue for evaluating the message. The work of Riggle et al. (1992), Ottati (1990), and Ottati et al. (1988; 1990) has also supported these findings. The research of Kuklinski and Hurley (1994) connected the use of ideological heuristics and source cues and, as a result, highlighted one area of possible peril when citizens draw on simple signals to formulate political evaluations. They argued that by focusing their attention on the individual political actor, citizens make quick judgments of the information presented to them based largely on the reputation of the speaker. They found that this was the case, as experimental subjects presented with a message evaluated that message based largely on their opinion of the speaker. An easy way to characterize this is that, because respondents are evaluating information based on their opinion about the speaker rather than an evaluation of the substantive content presented, the “messenger” is simply overwhelming the “message.” In this case, because both networks arguably send strong cues to the viewer regarding the ideological slant of their news coverage, the actual “message” either news network presents may be irrelevant so long as the viewer knows that the “messenger” was either CNN or FNC. I believe that the “messenger” may indeed be overwhelming the “message” due to viewers’ ideological perception of CNN and FNC. By assessing whether this is the case, we can hopefully draw firm conclusions as to whether viewers of television news are indeed drawing ideological cues from these networks, which, irrespective of story content, leads to an overall perception of ideological bias by viewers.

33 Hypotheses The two hypotheses for this examination are as follows:

Hypothesis One: Irrespective of the substantive content of news stories, attribution of those stories to either FNC or CNN will cause viewers to perceive the stories as ideologically biased.

I expect a significant difference in viewers’ perception of ideological bias between the networks, with news content attributed to FNC being perceived as conservatively biased, and news content attributed to CNN being perceived as liberally biased.

Hypothesis Two: Perceptions of ideological bias in news content will be most pronounced among ideologues whose views are at odds with those purportedly held by a given network.

The tendency to view news content attributed to FNC as conservatively biased should be most pronounced among liberals. Likewise, the tendency to view news content attributed to CNN as liberally biased should be most pronounced among conservatives. Methodology An experimental design was implemented to determine if the “messenger” was indeed overwhelming the “message” with regard to the substantive news content presented by both Fox News Channel and CNN. A group of 246 undergraduates recruited from various introductory and upper-level political science courses at Florida State University participated in this experiment. As detailed in Chapter two, this experiment involved respondents watching one of eight videotaped news broadcasts (Fox News Channel original, CNN original, Fox News Channel content attributed to either Fox News, CNN, or given no network attribution, and CNN content attributed to either Fox News Channel, CNN, or given no network attribution) and completing a survey instrument. The dependent variable for this analysis is the viewer’s perceived ideological bias of the news. This was measured in terms of how far away the viewer thought the story

34 was from an ideological neutral point. In this study respondents viewed the five stories and then were asked to rank them on a zero to ten scale, with zero representing strongly perceived liberal bias, five representing perceived ideological neutrality, and ten representing strongly perceived conservative bias. Because I am interested in the overall evaluation of bias in the news and not just bias in individual news stories, the scores of each story were summed in order to get a summary measure of ideological bias within the news program.17 Therefore, bias is measured on a zero to fifty scale, with zero representing absolute perceived liberal bias, fifty representing absolute perceived conservative bias, and twenty-five representing perceived ideological neutrality. There are several additional variables that need to be explained. The first of these is a measure of the overall perceived ideological bias of six major television news networks: CNN, FNC, ABC, CBS, MSNBC, and NBC. These variables were measured in the same manner as the ideological bias measure of the stories that were mentioned previously. Respondents were asked to rank the networks on a zero to ten scale, with zero representing strongly perceived liberal bias, five representing perceived ideological neutrality, and ten representing strongly perceived conservative bias. In the analysis it was necessary to control for the attribution of the stories to determine if this alone was driving bias scores. Two dichotomous variables were created to account for whether the stories were attributed to CNN or FNC. The CNN attribution variable is coded zero if the stories are not attributed to CNN and one if they are. Likewise, the FNC attribution variable is coded zero if the stories are not attributed to FNC and one if they are. Because hypothesis two deals exclusively with ideology, it was necessary to have a sound measure of that concept. Political ideology was initially measured on the traditional one (strong liberal) to seven (strong conservative) scale; this variable was recoded from negative three to three, with negative three signifying a “strong liberal”, zero signifying an “ideological moderate”, and three signifying a “strong conservative.” Because I am interested in how ideologues react to individual networks, it was necessary

17 A reliability analysis was conducted to determine if constructing a summary variable was appropriate. The summary scale was found to reach an acceptable level of reliability (Cronbach's alpha = 0.81). In addition, item-total correlations range from .405 to .523, which fall within the recommendation of between .30 and .70 put forth by Kerlinger (1986).

35 to create interaction terms. Political ideology was interacted with the FNC and CNN attribution dummy variables in order to determine how respondents harboring different ideologies respond to different network attributions. These two interaction terms allow for an assessment of differential ideological responses to news networks with allegedly different ideological viewpoints. It was also necessary to control for the original source of the news story to make sure that nothing about the content from one network or the other was driving bias scores. This is a dichotomous variable, which is coded zero if the original source of the content was FNC and one if the original source of the content was CNN. In addition, the original news story source variable was interacted with political ideology. Results The model in Table 3.1 serves as the key test of hypothesis one. To reiterate, I expect that news stories attributed to CNN will have the smallest mean on the ideological bias scale and that news stories attributed to Fox News Channel will have the largest, and that there will be a significant difference between the two. A regression model was utilized to investigate if there is a difference in means across the three network attributions. The findings presented in Table 3.1 provide support for my first hypothesis. Stories that are not attributed to either Fox News Channel or CNN register a score of 25.5 on my 0-50 ideological bias scale, meaning that people essentially perceive these stories to be ideologically neutral. The negative coefficient for CNN attribution confirms my suspicion that viewers perceive stories reported on that network as having a liberal bias, as stories attributed to CNN are seen as being roughly 4.3 points more liberal than stories that receive no network attribution. Likewise, the positive coefficient for news stories attributed to Fox News Channel confirms that viewers perceive stories presented on that network as having a conservative tilt, as stories attributed to this network are seen as being roughly 4.9 points more conservative than stories that received no network attribution. The fact that both the Fox News Channel and CNN attribution variables are significant confirms that each of these attributions impacts respondents’ rating of the stories presented. The insignificant content variable illustrates that there is no perception of an underlying ideological bias within the content of the stories presented by either Fox

36 News or CNN. Whether the original source is Fox News Channel or CNN, respondents are only responding to the attribution when formulating opinions regarding ideological bias within news stories. The model presented in Table 3.2 presents the key test for hypothesis two. 18 I expect conservative viewers to perceive news stories attributed to CNN as being much more liberal than do moderates or liberals. Likewise, I anticipate that liberal viewers will perceive news stories attributed to Fox News Channel as being much more conservative than do moderates or conservatives. A regression analysis was also utilized to investigate this hypothesis. Again, the results comport well with expectations. The following equations isolate the conditional nature of the relationship between network attribution and political ideology. The influence with regard to news content attributed to CNN is illustrated in Equation 1: EQ 1: Influence of CNN Attribution = [-4.549 – 1.501 (Ideology)] The significant coefficient of -4.549 operating on CNN Attribution is the conditional effect for ideological moderates. What this reveals is that merely attributing news content to CNN cues ideological moderates to a liberal news bias, as this attribution moves this ideological group 4.549 points in a negative direction on the ideological bias scale. The significant interaction indicates that ideology conditions the influence of attribution. If you will recall, the coding of liberals ranged from -3 to -1. Therefore, as respondents become increasingly liberal, the effect of attributing news content to CNN becomes less influential, ultimately reaching a point at which attribution to CNN has essentially no influence on strong liberals (i.e., strong liberals do not perceive ideological bias from content attributed to this network). Coding of conservatives, on the other hand, ranged from 1 to 3. As a result, as respondents become increasingly conservative, the effect of attributing news content to CNN is exacerbated, as strong conservatives are more apt than any other ideological group to detect liberal bias in news content attributed to CNN.

18 I ran additional models that included control variables for media use patterns (i.e. network preference, hours per week watching TV news, propensity to read the newspaper, propensity to visit online sites, newspaper and/or website used most frequently) as well as three-way interactions between ideology, content, and network attribution. The variables failed to reach significance and did not make a significant contribution to the model. As a result, they were dropped from the final analysis.

37 The conditional relationship with regard to news content attributed to Fox News Channel is illustrated in Equation 2: EQ 2: Influence of Fox News Channel Attribution = [4.736 – 1.502 (Ideology)] Once again, the significant coefficient of 4.736 operating on Fox News Channel attribution is the conditional effect for ideological moderates. This indicates that simply attributing news content to Fox News cues ideological moderates to a conservative news bias, as this attribution moves this ideological group 4.736 points in a positive direction on the ideological bias scale. In addition, the significant interaction indicates that ideology conditions the influence of Fox News Channel attribution. In this case, as respondents become increasingly liberal, the influence of attributing news content to Fox News Channel reaches its zenith, as strong liberals are most likely of all ideological groups to detect ideological bias in news content attributed to this network. In contrast, for strong conservatives, attributing news content to Fox News Channel has little substantive effect. Figure 3.1 demonstrates the substantive effect of these interaction terms. In this Figure, the bias score is calculated for respondents who viewed news content that originated from CNN.19 The three sets of bars represent, from left to right, the perceived bias scores for content attributed to CNN, content given no network attribution, and content attributed to Fox News. The bar filled with horizontal lines signifies perceived ideological bias among strong liberals, the checked bar represents perceived ideological bias among political moderates, and the bar filled with vertical lines represents perceived ideological bias among strong conservatives. As the figure illustrates, for a strong liberal, the CNN attribution conveys nothing extraordinary, and thus strong liberals’ perceived ideological bias scores are almost identical for CNN attribution and for the no attribution control. However, mere attribution of a story to Fox News Channel brings a stark nine-point jump on the dependent variable, with strong liberals seeing the Fox News Channel label as a signal of conservative bias. Precisely the opposite pattern emerges for strong conservatives. For

19 CNN content was used in the analysis presented in Figure 1. FNC content could have been used, but given the small coefficients and the high levels of insignificance associated with the content variable and the interaction between content and ideology, no significant substantive difference would have emerged by creating the figure using content presented by the other network.

38 strong conservatives, attributing a story to Fox News Channel brings no change in perceptions of ideological bias relative to the no attribution baseline, but perceived bias falls nine points when coverage is linked with CNN. It is when we view the CNN and Fox News effects in tandem that the impact of the ideological signals transmitted by the respective networks becomes most vivid. Strong conservatives give content attributed to CNN a rating of 13.864. Strong liberals give the very same content attributed to Fox News a rating of 35.980. Holding actual content constant, the mere attribution of that content to one network or the other causes an enormous disparity among strong liberals and conservatives; a disparity which, at over 22 points, spans nearly half of the range of the story ideology scale. In contrast to ideologues, who perceive only networks with ideological reputations at odds with their own as biased, Figure 3.1 reveals that the CNN and Fox News Channel attributions both matter for ideological moderates. When a network’s reputation is not at odds with a respondent’s ideological preferences, bias is not perceived; otherwise, the mere linking of a story to Fox News Channel or CNN sends a powerful ideological signal. Conclusion and Implications Ideological bias in television news is one of the more discussed, yet least understood, phenomena for observers of contemporary political science. My research has attempted to shed light on one important facet of this topic. I have argued that attaching the “CNN” and “Fox News Channel” labels to news stories sends an ideological signal to the viewer, which could lead the viewer to perceive bias and to raise cognitive roadblocks that prevent the proper acquisition and processing of the information. My findings have largely supported this argument; at least in the case of CNN and Fox News Channel, the messenger does appear to be overwhelming the message. Utilizing an experimental approach permitted me to hold news content constant, while manipulating network attribution. My results demonstrate that people do react to the label of the news organization to which the story is attributed. Irrespective of the actual content of the stories presented, the CNN and Fox News Channel labels send a clear signal to the viewer that the content of the reports is ideologically biased: to the liberal side for CNN and to the conservative side for Fox News.

39 Furthermore, these perceptions of bias are most pronounced among ideologues who believe that their views are most at odds with those of a network. Perceptions of liberal bias in news content attributed to CNN are most pronounced among strong conservatives, and perceptions of conservative bias in news content attributed to Fox News Channel are most pronounced among strong liberals. That there is significant difference between the two networks across these two ideological groups was not surprising, as one would expect them to be more attuned to particular examples of language that may indicate ideological bias. In addition, I was also able to demonstrate that these perceptions of bias were not lost among weaker ideologues, as even moderate viewers identified a liberal bias in broadcasts attributed to CNN and a conservative bias in broadcasts attributed to Fox News Channel. What are the potential implications of the findings? First, it is clear that the perceptions of CNN and Fox News Channel as being ideologically biased are pervasive in society. In this chapter I have clearly illustrated that individuals can watch virtually identical stories on CNN and Fox News yet still derive different ideological signals from each network. These findings would appear to provide support for those on both sides of the aisle who allege that the media has either a conservative (Alterman 2003; Miller 2002; Gitlin 2003) or liberal (Goldberg 2001; Maitre 1994) bias. However, the failure of respondents to identify media bias in instances where the news stories are not given a network attribution draws these allegations into question. Perhaps a more apt description of what occurs is that media bias is merely an artificial construction that has become so ingrained in American culture that the organization presenting the news has become more important than the product it presents. This is problematic because it could result in people dismissing unbiased news reports because they wrongly believe that they are biased. In the worst case, the misconception that CNN and Fox News function as the news outlets of the liberal and conservative viewpoint, respectively, could ultimately lead to public confusion regarding what political ideals liberals and conservatives actually strive to uphold. These findings also lend support to scholars who warn of the potential dangers of reliance upon heuristic shortcuts. At first glance one could argue that these findings support the notion, first put forward by Converse (1964), that the citizenry is woefully

40 ignorant with regard to politics and, in this case, political news. However, respondents were able to identify apparently neutral news reports as being neutral when not presented with a network attribution, which indicates that the citizenry does possess the capacity to perform an in-depth assessment of information presented to them on a news broadcast. The problem emerges when citizens are also presented with the network attribution. This allows respondents to simplify and engage in peripheral processing, and in this case the reputation of these networks appears to lead citizens to have misperceptions about the ideological nature of the information presented. These misperceptions, or mistakes, in the evaluation of information that result when citizens rely on cognitive heuristics, is the “cautionary tale” that was told by Kuklinski and Hurley (1994) and supported by other scholars in both psychology (Tversky and Kahneman 1974) and political science (Lau and Redlawsk 1997; Mondak 1994). The results presented here have interesting implications for other areas of research in the field of media and politics. Although nothing in my findings rejects the significance of phenomena such as priming, framing, and agenda setting, my results do indicate that those effects may vary widely across individuals depending on what type of ideological bias, if any, each person perceives in the news. As Miller and Krosnick (2000) pointed out, perceptions of bias translate into media distrust, and can possibly prevent the news media from successfully fulfilling its priming and agenda setting functions. Walsh (2003) makes a related point when she argues that media effects interact with individuals’ underlying political perspectives. If nothing else, future researchers in these areas can gain added nuance by recognizing that, for example, the capacity of CNN to produce an agenda-setting effect may well depend in part on the ideological leaning of the particular viewer. Although viewers can be assigned to particular viewing conditions in the laboratory, self-selection operates in the real world. One manifestation of self-selection is that some liberals avoid watching Fox News Channel, and some conservatives steer clear of CNN. These patterns warrant assessment in light of the current findings. It is certainly true that Fox News has a more conservative viewership and that liberals are more likely to watch CNN, but it is not the case that liberals and conservatives fully abstain from Fox News and CNN, respectively. Therefore, while self-selection may limit

41 the effects on the margins, it certainly does not eliminate the effects. Furthermore, these individuals are self-selecting because they think they disagree with the network, as earlier I demonstrated that, absent network attribution, the news content presented in this experiment was perceived as being ideologically neutral. If anything, this adds another layer of support to my argument that network reputation, and only network reputation, matters with regard to perceptions of ideological bias in television news. The obvious next step in this research agenda is to examine what, if any, impact these perceptions of bias have on how the American citizen processes television news. Specifically, research needs to examine whether individuals are better able to acquire and retain information presented to them by a network they believe to be consonant with their ideological beliefs or by one that they believe to be ideologically dissonant. Compelling arguments can be made to support both theses. It seems plausible that someone would remember more information presented by a consonant media source, as it would serve to reinforce his or her existing beliefs. However, it is also plausible that people will remember more information presented to them by a dissonant source, because the sheer fact that they disagree with the source may serve as a factor motivating them to cognitively engage the material. This question is examined in the next chapter.

42 Table 3.1 Perceived Ideological Bias of Coverage by Experimental Treatment Group

B SE P-value

Constant 25.509* .730 .000

CNN Content -.323 .705 .647

CNN Attribution -4.314* .872 .000

FNC Attribution 4.946* .880 .000

N = 218 F = 44.24 Prob>F = .0000 Adj. R2 = 0.374 *p<.001.

43 Table 3.2 Ideology and Perception of Bias Among Fox News Channel and CNN

B SE P-value

Constant 25.392 .626 .000

CNN Content -.529 .605 .384

CNN Attribution -4.549* .756 .000

FNC Attribution 4.736* .753 .000

Ideology -.416 .404 .305

Ideology X CNN Attribution -1.501* .501 .004

Ideology X FNC Attribution -1.502* .514 .001

Ideology X CNN Content -.209 .432 .629

N = 216 F = 36.514 Prob>F = .0000 Adj. R2 = 0.535 *p<.05

Strong Liberal 30 Moderate

20 Strong Conservative

0 CNN N/AN/A Fo x FNC New s

Figure 3.1 Ideology and Perceptions of Ideological Bias in News by Network Attribution

CHAPTER FOUR

MEMORABLE ENCOUNTERS: IDEOLOGY, INFORMATION ACQUISITION, AND TELEVISION NEWS

Millions of Americans get their news from America’s two leading cable news networks, Fox News Channel and CNN. If you recall, in the previous chapter I demonstrated that citizens perceive Fox News Channel and CNN as being conservatively and liberally biased, respectively, and that these network attributions send many viewers a clear signal regarding the ideological nature of the news content they present. As a result, one might assume that, in order to achieve cognitive expediency and limit cognitive dissonance, only conservative viewers would watch Fox News and only liberal viewers would tune in to CNN. However, this is not the case, as previous research has demonstrated that more than a third of Fox News Channel’s audience do not identify themselves as conservative, and roughly two-thirds of CNN’s audience do not self- identify as liberal.20 Given the perceptions regarding the ideological nature of the news content that CNN and Fox News Channel present, the fact that these networks do indeed have such ideologically diverse audiences raises a number of interesting questions regarding how individuals react to viewing news from what they perceive to be dissonant news sources. One such question is pursued in this chapter: whether individuals are better able to acquire and retain information presented to them by a news network that they believe to be ideologically consonant or by one that they believe to be ideologically dissonant with their own individual beliefs. This question is important because disseminating information to the public is the primary job of the news media, given their role as

20 Poll conducted by Garin-Hart-Yang Research Group in November 2004.

46 political informant of the American people. Likewise, acquiring information is a, or more likely the, central reason that people follow the news. Therefore, the extent to which citizens acquire and retain information presented to them on the news speaks directly to how well this vital information system is functioning. Compelling arguments can be made to support both theses. It seems plausible that individuals would remember more information presented to them from an ideologically consonant source, as such information is easier to process, and the whole logic of cognitive sophistication is that the more people know about a subject, the easier it is for them to assimilate new information. In addition, this consonant information would serve to reinforce their existing political beliefs. However, it is also plausible that individuals remember more information presented to them from a dissonant source because the sheer fact that the source is dissonant may lead them to be cognitively engaged, on guard, and ready to counterargue. This chapter examines the impact of the ideological consonance or dissonance of a television news source on an individual’s ability to acquire and recall information. I will begin with a review of the existing literature that has touched on the influence of consonant and dissonant information sources. I will then provide a discussion of the experimental design and measurement techniques utilized for this undertaking, report the results of my hypothesis tests, and conclude with a discussion of the theoretical and practical implications of these findings. Theoretical Background In order for the results of this chapter to be of much substantive importance, it is essential that more than a trivial percentage of television news viewers expose themselves to dissonant news sources in the real world. Examinations of selectivity theory have demonstrated that a sizable amount of news outlet selectivity does take place (Katz 1981; Klapper 1960; McGuire 1985; Axelrod 1973). These scholars argue that individuals selectively expose themselves to like-minded media sources and content. They contend that this takes place because individuals like to screen out information that they do not like or that challenges their fundamental social or political values. However, the issue of selective media exposure has yet to be completely resolved, as several laboratory experiments have failed to conclusively demonstrate that individuals choose to expose

47 themselves to politically consonant media sources (Freedman and Sears 1965; Frey 1986; Sweeney and Gruber 1984; Zillman and Bryant 1985). In addition, recent polling data have demonstrated that the audiences for both Fox News Channel and CNN are more ideologically diverse than would be expected.21 As a result, it is clear that, although some selectivity does occur, a significant number of individuals are exposing themselves to what they perceive to be dissonant media outlets. Having established that people do indeed expose themselves to ideologically dissonant media sources, the next question is whether this dissonance will function as a positive or negative force with regard to information acquisition and retention. The majority of this research supports the notion that dissonance should exert a negative influence with regard to information acquisition and retention. Reputable scholars conducting research on the congeniality hypothesis have demonstrated that information individuals perceive to be ideologically dissonant with their beliefs is much less memorable than information individuals believe to be ideologically consonant with their beliefs. For example, in an experiment in which pro-communist and anti-communist respondents were presented with information that was pro-Soviet Union and anti-Soviet Union, the pro-communist group remembered more pro-Soviet Union information, and the anti-communist group remembered more anti-Soviet Union information (Levine and Murphy 1943). These findings regarding the congeniality hypothesis were also supported by early work done by Edwards (1941). In addition, research on cognitive dissonance (Festinger 1957) supports the notion that individuals should be better able to acquire and retain information presented to them by consonant media sources. Cognitive dissonance refers to the tension that results when individuals hold dissonant cognitions, or thoughts that contradict each other. The theory states that the mental tension resulting from contradicting cognitions motivates individuals to reduce their amount of dissonance between cognitions. The reduction of dissonance can be achieved in a number of ways; individuals can dismiss or ignore the information with which they are presented, add new cognitions to reinforce their initial position, or alter their cognitions in order to bring them into consonance. Because the average American citizen is perhaps best described as a cognitive miser, the likelihood is

21 Poll conducted by Garin-Hart-Yang Research Group in November 2004.

48 that most will take the path of least resistance in reducing their individual level of dissonance. This reduction is most readily achieved by either dismissing, or completely ignoring, the information presented by sources perceived to be dissonant. This idea is supported by those who question whether learning takes place as a result of citizens’ encountering of dissonant political information (Gastil and Dillard 1999; Cook et al. 1999). Proponents of balance theory (Heider 1946; Newcomb 1953) would also support the idea that individuals are more apt to remember information presented to them by a consonant media source. Balance theory is a motivational theory of attitude formation wherein the drive toward consistency is the motive for psychological balance. Balance is achieved when triadic relationships are viewed as being multiplicatively positive (A likes B, A likes C, and B likes C; or A likes B, A dislikes C, B dislikes C). Because psychological balance would be most easily achieved by accepting consonant information and dismissing dissonant information, this theory would predict that dissonance would have a negative influence on memory. In examining how individuals deal with schema-consistent and schema- inconsistent information, Fiske and Taylor (1991) found that individuals were more likely to process schema-consistent than schema-inconsistent information. Research done on racial stereotypes also supports this notion. Peffley and Hurwitz (1997) demonstrated that these stereotypes played a significant role in biasing respondents’ perceptions of welfare recipients and as a result their evaluations of political programs designed to assist them. The authors found that whites who held negative stereotypes about African- Americans, oftentimes even in the face of counter-stereotypical information, judged African-American program targets more harshly than similarly described white program targets. In addition, the authors assert that, although possible, the likelihood of changing these perceptions is quite small. The same logic could be applied to television news. If viewers believe a news network is biased, they are likely to maintain that belief, and thus not give much consideration to the inconsistent information a news network presents. Finally, experimental work on the hostile media phenomenon (Vallone, Rose, and Lepper 1985) supports the notion that individuals are more likely to acquire and recall information from a consonant source. When presented with the same relevant news

49 story, both pro-Israeli and pro-Arab groups thought that the story was biased against their political position. However, the majority of these respondents were unable to provide specific factual examples as to why they thought the stories were biased. Therefore, it appears that these respondents were prematurely dismissing, or perhaps altogether ignoring, the information from the broadcast because they had identified the provider of the information as a dissonant source. Although a large amount of research supports the idea that viewers should be less apt to acquire and recall information presented by a dissonant news source, the issue is far from settled. Equally reputable scholars examining the congeniality hypothesis have brought this theory into question. Later studies of information acquisition have been unable to replicate the earlier findings regarding the congeniality hypothesis. Some scholars were only able to produce null effects (Greenwald and Sakumura 1967), while some demonstrated that the circumstances surrounding the reception of congenial information dictated whether it was more memorable (Jones and Aneshansel 1956). More recent attempts have actually demonstrated that uncongenial information was more memorable than congenial information because people actually engaged the dissonant information in an in-depth manner (Petty and Cacioppo 1979; Chaiken et al. 1989; Edwards and Smith 1996; Ditto and Lopez 1992; Ditto et al. 1998). In addition, many scholars would argue that a sizable segment of the population turns to the mass media specifically to obtain politically dissonant information -- because the media are in fact “hostile” to their position. Research on social networks and cross- cutting information has shown that individuals strive to avoid confrontation in their life, which leads them to choose politically homogenous discussion partners (Mutz 2006; Bennet, Fisher, and Resnick 1994; Ulbig and Funk 1999; Huckfeldt, Johnson, and Sprague 2004). However, the same political homogeneity is not as prevalent with regard to citizens’ choice of news media sources, as they appear to be more willing to expose themselves to dissonant news sources (Mutz 2002; Mutz and Martin 2001). Some research suggests that this exposure to ideologically dissimilar views does not discourage the accommodation of the new information (Lane 1962) and, in fact, may encourage greater personal deliberation and reflection about one’s views (Arendt 1968; Habermas 1989).

50 This work has also received support from social psychologists examining controversy versus concurrence (Lowry and Johnson 1981; Smith et al 1981). According to these scholars, controversy exists when one individual’s information or ideas are at odds with those of another individual (or other individuals). Concurrence occurs when individuals seek to avoid this information in order to reach an agreement on a specific topic. In an experimental study using small groups outside of the realm of political decision making, these scholars found that an emphasis on controversy over concurrence- seeking promotes greater mastery and retention of information. Hypothesis Although the matter is certainly not resolved, given the findings of the majority of previous studies I anticipate that dissonance will exert a negative influence on the acquisition and retention of information. As a result, the hypothesis for this examination is as follows:

Hypothesis 1: Respondents will acquire and recall more information from news networks that are ideologically consonant with their political beliefs.

Strong conservatives and strong liberals should be more apt to acquire and retain information presented by Fox News Channel and CNN, respectively, because such information is ideologically consonant with their political beliefs. On the other hand, strong conservatives and strong liberals should be more apt to dismiss information from CNN and Fox News Channel, respectively, because these news sources are perceived to be ideologically dissonant. Moderates should be more apt to retain news content presented with no network attribution, because both CNN and Fox News Channel are dissonant with their political stances. Methodology An experimental design was implemented to determine if individuals were better able to acquire and recall information from news broadcasts attributed to ideologically consonant media sources or from broadcasts attributed to ideologically dissonant media sources. A group of 246 undergraduates recruited from various introductory and upper- level political science courses at Florida State University participated in this experiment.

51 As detailed in chapter two, this experiment involved respondents watching one of eight videotaped news broadcasts (Fox News Channel original, CNN original, Fox News Channel content attributed to either Fox News, CNN, or given no network attribution, and CNN content attributed to either Fox News Channel, CNN, or given no network attribution) and completing a survey instrument. There are three dependent variables utilized in this analysis. The first of these is a recall attempts variable. This was measured in terms of how many specific facts the respondents attempted to recall from the news stories they had previously viewed. In this study respondents viewed five news stories and then were asked to recall five specific facts from two of the stories: the story on the war in Iraq and the story on the female suicide bomber in Israel containing CNN content, and the story on the war in Iraq and the story on mad cow disease featuring Fox News content.22 Each time a respondent put an answer in one of the allotted spaces, they were credited with a recall attempt. Because I am interested in the respondents’ overall ability to recall facts presented on both networks, the variables were summed in order to get an overall picture of the respondents’ recall ability.23 Therefore, perceived recall is measured on a 0-10 scale, with 0 signifying that the respondent did not attempt to recall any specific facts and 10 signifying that the respondent attempted to recall 10 specific facts. This variable is referred to in the text as “total recall attempts.” The second dependent variable indicates the ability of respondents to correctly recall information presented to them in the news. This was measured in terms of how many specific facts respondents were able to correctly recall from the news stories they had previously viewed. This variable was coded by comparing the factual recall attempts of the respondents to the transcripts of the content within the individual stories. If the fact that the respondent was able to recall was present within the transcript of the story, the recall attempt was coded as being correct. This variable is measured on a 0-10 scale, with 0 signifying that the respondent could not recall any specific facts correctly and 10 signifying that the respondent was able to recall 10 specific facts correctly. This variable is referred to in the text as “correct recall.”

22 The two stories selected for this examination were randomly chosen from the five stories presented by each network. 23 Cronbach’s alpha = .69

52 The final dependent variable measures the propensity of respondents to be incorrect in their attempts to recall factual information from news stories. This was measured in terms of how many of the specific facts that respondents attempted to recall from the news stories were incorrect. This variable was also coded by comparing the factual recall attempts of the respondents to the transcripts of the content within the individual stories. If the fact that the respondent was able to recall was not present within the transcript of the story, the recall attempt was coded as being factually incorrect. This variable is measured on a 0-10 scale, with 0 signifying that none of the facts recalled by a respondent was incorrect and 10 signifying that all ten of the facts recalled by a respondent were incorrect. This variable is referred to in the text as “incorrect recall.” With regard to the independent variables of interest, it was first necessary to account for the attribution of the stories. Two dichotomous variables were created to account for whether the stories were attributed to CNN or Fox News Channel. CNN attribution is coded 0 if the stories are not attributed to CNN, and 1 if they are. Likewise, Fox News Channel attribution is coded 0 if the stories are not attributed to Fox News, and 1 if they are. Political ideology was initially measured on the traditional 1 to 7 scale; this variable was recoded from -3 to 3, with -3 signifying “strong liberal” and 3 signifying “strong conservative.” Because I am interested in differences in how ideologues process information presented by ideologically consonant and dissonant networks, it was necessary to create interaction terms. This was done by interacting political ideology with both the CNN attribution and the Fox News Channel attribution variables. These interaction terms will allow for a comparison of information acquisition and recall among those watching consonant and dissonant media sources. It was also necessary to control for the original source of the story to make sure that the content from one network was not more memorable than content from the other network. This was accomplished via a dummy variable coded 0 if the original source of the content was Fox News Channel and 1 if the original source of the content was CNN. Finally, the Fox News and CNN attribution variables were interacted with the original source variable.

53 Results The model in table 4.1 tests whether respondents make more factual recall attempts when they are presented with information from a dissonant news media source than when they are presented with information from a consonant media source. Because the dependent variables are counts, a Poisson regression was utilized to investigate these hypotheses.24 Surprisingly, the findings presented in Table 4.1 support the notion that respondents better acquire and retain information presented to them by a network that they should find ideologically dissonant.25 The first column in table 4.1 presents the results regarding recall attempts, which has a mean of 7.57 and a standard deviation of 2.19. The following equations isolate the conditional nature of the influence of network attribution and political ideology on recall attempts. The influence with regard to news content attributed to CNN is illustrated in Equation 1: EQ 1: Influence of CNN Attribution = [.128 + .056 (Ideology)] The significant coefficient of .128 operating on CNN Attribution is the conditional effect for ideological moderates. This positive coefficient indicates that merely attributing news content to CNN inspires more factual recall attempts from this ideological group in comparison to news content that receives no network attribution. The significant interaction indicates that ideology conditions the influence of attribution. If you will recall, the coding of liberals ranged from -3 to -1. Therefore, as respondents become increasingly liberal, the effect of attributing news content to CNN becomes less influential, ultimately reaching a point to where attribution to CNN has essentially no influence on strong liberals (i.e., strong liberals do not make more factual recall attempts when they are exposed to news content attributed to this network as opposed to when they are exposed to news content given no network attribution). In contrast, the coding of conservatives ranged from 1 to 3. As a result, as respondents become increasingly conservative, the effect of attributing news content to CNN is exacerbated, as strong

24 Tests for dispersion indicated that a Poisson regression was appropriate for this examination. 25 Additional models that included interactions between content and ideology, as well as three-way interactions between content, ideology, and attribution, were also investigated. However, these variables made no significant contribution to the model, and thus were not reported here.

54 conservatives make more factual recall attempts when they are presented with news content attributed to CNN than does any other ideological group. The conditional relationship with regard to news content attributed to Fox News is illustrated in Equation 2: EQ 2: Influence of Fox News Channel Attribution = [.166 - .099 (Ideology)] Once again, the significant coefficient of .166 operating on Fox News Channel attribution is the conditional effect for ideological moderates. This positive coefficient indicates that merely attributing news content to Fox News Channel inspires more factual recall attempts from this ideological group in comparison to news content that receives no network attribution. Ideology once again conditions the influence of Fox News attribution. In this instance, as respondents become increasingly liberal, the influence of attributing news content to Fox News Channel reaches its pinnacle, as strong liberals are the ideological group that makes the most factual recall attempts when news content is attributed to Fox News. In contrast, for strong conservatives, attributing news content to Fox News Channel has little substantive effect on factual recall attempts, when compared to their number of attempts when information is given no network attribution. The coefficient for the content variable is substantively very small and highly insignificant. This illustrates that, among moderates, news content that originates from Fox News Channel does not inspire more recall attempts than content originating from CNN, and vice versa. In addition, the coefficient for the interaction between political ideology and original source of the news content is small and highly insignificant, which demonstrates that Fox News also does not inspire more recall attempts among ideologues than does content originating from CNN, and vice versa. The model in table two tests whether respondents make more factually correct recall attempts when they are presented with news content from a dissonant news media source or when they are presented with news content presented by a consonant media source. This variable has a mean of 4.71 and a standard deviation of 2.60. As was the case with recall attempts, because the dependent variables are counts, a Poisson regression was utilized to investigate these hypotheses.26

26 Tests for dispersion indicated that a Poisson regression was appropriate for this examination.

55 Once again, the results presented in Table 4.2 indicate that respondents are more likely to correctly recall information from broadcasts attributed to an ideologically dissonant news network. The influence with regard to news content attributed to CNN is illustrated in Equation 3: EQ 3: Influence of CNN Attribution = [.318 + .136 (Ideology)] The significant coefficient of .318 operating on CNN Attribution is the conditional effect for ideological moderates. This positive coefficient demonstrates that attributing news content to CNN results in more correct factual recall attempts from this ideological group than when its members are exposed to news content that receives no network attribution. The significant interaction indicates that ideology conditions the influence of attribution. As respondents become increasingly liberal, the effect of CNN attribution is essentially neutralized, as strong liberals are no more apt to correctly recall information presented by CNN than they are to correctly recall information given no network attribution. However, as respondents become increasingly conservative, the effect of attributing news content to CNN is very influential, as strong conservatives make more correct factual recall attempts when they are presented with news content attributed to CNN than does any other ideological group. The conditional relationship with regard to news content attributed to Fox News Channel is illustrated in Equation 4: EQ 4: Influence of Fox News Channel Attribution = [.392 - .194 (Ideology)] The significant coefficient of .392 operating on Fox News Channel attribution is the conditional effect for moderates. The fact that this coefficient is positive demonstrates that attributing news content to Fox News inspires more factual recall attempts from this ideological group than when its members are exposed to news content that receives no network attribution. The significant interaction indicates that ideology conditions the influence of Fox News Channel attribution. In this instance, as respondents become increasingly liberal, the influence of attributing news content to Fox News Channel reaches its pinnacle, as strong liberals are the most likely of all ideological groups to make correct factual recall attempts when exposed to news content attributed to Fox News. In contrast, attributing news content to Fox News Channel has little substantive effect on correct factual recall attempts among conservatives, as there is no

56 significant difference between the number of correct recall attempts made when members of this group are exposed to news content attributed to Fox News Channel and when they are exposed to news content given no attribution. Once again, the coefficients for both the content variable, and the interaction between content and political ideology, are small and statistically insignificant. This illustrates that, among moderates and stronger ideologues alike, news content that originates from Fox News does not inspire more correct recall attempts than does content originating from CNN, and vice versa. The statistical model in table 4.3 presents the results regarding incorrect recall attempts, which has a mean of 2.87 and a standard deviation of 1.64. In this case, respondents were less apt to be incorrect in their recall attempts when presented with information by a news network they perceive to be ideologically dissonant. Once again, a conditional relationship exists between network attribution and political ideology with regard to incorrect recall attempts. The influence with regard to news content attributed to CNN is illustrated in Equation 5: EQ 5: Influence of CNN Attribution = [-.127 - .070 (Ideology)] The significant coefficient of -.127 operating on CNN Attribution is the conditional effect for ideological moderates. In this case the coefficient is negative, which demonstrates that attributing news content to CNN results in fewer incorrect factual recall attempts from this ideological group in comparison to when its members are exposed to news content that receives no network attribution. The significant interaction indicates that ideology conditions the influence of attribution. As respondents become increasingly liberal, the effect of CNN attribution is exacerbated, as strong liberals are more likely to make incorrect recall attempts when they are exposed to information presented by CNN than when they are exposed to news content given no network attribution. However, as respondents become increasingly conservative, the effect of attributing news content to CNN has the opposite effect, as strong conservatives make fewer incorrect recall attempts when news content is attributed to CNN as opposed to when content is given no attribution. The conditional relationship with regard to news content attributed to Fox News Channel is illustrated in Equation 6:

57 EQ 6: Influence of Fox News Channel Attribution = [-.206 + .075 (Ideology)] The significant coefficient of -.206 operating on Fox News Channel Attribution is the conditional effect for ideological moderates. As was the case with CNN attribution, the coefficient is negative, which demonstrates that attributing news content to Fox News Channel results in fewer incorrect factual recall attempts from this ideological group in comparison to when its members are exposed to news content that receives no network attribution. The significant interaction indicates that ideology conditions the influence of attribution. As respondents become increasingly conservative, the effect of Fox News attribution is magnified, as strong conservatives are more likely to make incorrect recall attempts when information is presented by Fox News Channel than when information is given no network attribution. However, as respondents become increasingly liberal, the effect of attributing news content to CNN has the opposite effect, as strong liberals make fewer incorrect recall attempts when news content is attributed to CNN as opposed to when content is given no attribution. Once again, the coefficients for the content variable and the interaction between content and political ideology are small and statistically insignificant. This illustrates that, among moderates and stronger ideologues alike, news content that originates from Fox News Channel does not inspire more incorrect recall attempts than does content originating from CNN, and vice versa. Poisson regression coefficients are informative in regard to the direction of the effect of independent variables and in regard to the statistical significance of these variables. However, they are difficult to interpret and provide little information regarding the substantive impact of each independent variable. Therefore, predicted counts were calculated in order to highlight the actual difference in recall attempts, correct answers, and incorrect answers among strong ideologues. I will first examine the influence of network consonance or dissonance on the acquisition and retention efforts of strong liberals. First, as illustrated in Figure 4.1, there is no great difference in number of recall attempts when content is attributed to either CNN or given no attribution, as liberals make an average of 5.85 and 6.08 recall efforts, respectively. However, there is a large substantive effect when this same content is attributed to Fox News Channel, as under this circumstance liberals make an average of

58 9.67 recall attempts, which is roughly three and a half more attempts than when the content is given no attribution, and nearly four attempts more than when the content is attributed to CNN. With regard to correct recall attempts, there is no tangible difference between the number of correct responses offered by strong liberals when content is attributed to CNN (2.90 correct responses) or given no attribution (3.18 correct responses). However, when the news content is attributed to Fox News Channel, the number of correct responses increases by more than five to an average of 8.41 correct responses. With regard to incorrect responses there is once again little difference between liberals watching CNN and liberals watching content given no attribution, as these respondents were wrong an average 2.95 and 2.90 times, respectively. Yet, when the news content is attributed to Fox News Channel, the number of incorrect responses declines dramatically to a mere 1.26. To summarize, liberals attempt to recall more facts, more of those recall attempts are correct, and fewer of those recall efforts are incorrect, when content is attributed to Fox News Channel as opposed to being attributed to CNN or given no network attribution. A similar pattern emerges with regard to political conservatives. First, with regard to number of recall attempts, little substantive difference emerges between news content attributed to Fox News Channel (6.34) and news content given no network attribution (7.24). However, there is a notable substantive effect when this same news content is attributed to CNN, as the number of recall attempts made by conservatives increases to 9.73, which is nearly two attempts more than when the content is attributed to Fox News Channel, and nearly two and a half attempts more than when the content is given no network attribution. With regard to correct recall attempts, once again no major substantive difference emerges among conservatives when news content is attributed to Fox News and when content is given no network attribution. Conservatives make 2.99 correct attempts in the former scenario, and 3.61 correct attempts in the latter. Yet, when the same news content is attributed to CNN, the predicted number of correct responses is 7.46, which is more than a twofold increase over both of the other network attributions. In terms of incorrect recall attempts, conservatives watching content attributed to Fox News Channel and

59 given no attribution make roughly the same number of incorrect recall attempts, 3.35 and 3.63, respectively. However, when the same news content is attributed to CNN, the predicted number of incorrect recall attempts declines to 2.27. Therefore, strong conservatives attempt to recall more facts, more of those recall attempts are correct, and fewer of those recall efforts are incorrect, when content is attributed to CNN as opposed to being attributed to Fox News Channel or being given no network attribution. Conclusion and Implications The acquisition and retention of political knowledge is one of the most important topics of discussion for practitioners of contemporary political science. My research has attempted to shed light on one important facet of this topic. Investigating the influence of consonant and dissonant media sources on information acquisition and retention has allowed me to demonstrate that, surprisingly, respondents who are exposed to an ideologically dissonant media source will make more recall attempts, be correct more often, and be incorrect less often than those exposed to an ideologically consonant media source. Simply put, with regard to information acquisition and retention, respondents appear to be better off watching an ideologically dissonant network than an ideologically consonant one. Utilizing an experimental approach permitted me to hold news content constant, while providing ideologically dissonant and consonant network attributions. My results demonstrate that the perceived consonance or dissonance of a source does matter with regard to information acquisition and retention. Self-identified liberals and conservatives make significantly more recall attempts, are correct more often, and incorrect less often when watching Fox News Channel and CNN, respectively, than when watching the network that they believe to be more suited to their ideological predispositions. What are the potential implications of these findings? First, these results bring into question the notion, supported by proponents of cognitive dissonance (Festinger 1957) and balance theory (Heider 1946; Newcomb 1953), as well as those who have examined the processing of schema-consistent and schema-inconsistent information (Fiske and Taylor 1991), that individuals simply take the path of least cognitive resistance when processing information. In this case respondents, even when presented with the exact same news content, acquire and retain more information when that content

60 is presented by a news network that transmits an ideologically dissonant political signal. This is clearly contrary to what would be expected, which is that respondents would simply dismiss information that is attributed to a dissonant source. These results also raise the possibility that ideologically dissonant sources do a better job of providing “memorable” information specifically because they are perceived, or assumed, to be dissonant. Believing that a source of information is ideologically dissonant appears to activate a cognitive tendency that inspires individuals to pay more attention (Lane 1962; Arendt 1968; Habermas 1989; Lowry and Johnson 1981; Smith et al 1981). Perhaps the respondent wants to be able to formulate a counterargument as to why the position this news source is taking is incorrect. The old cliché “keep your friends close and your enemies closer” may apply here. Perhaps ideologues are already familiar enough with what their side thinks that they do not pay much attention to information provided by consonant news sources. Yet, these same individuals may want to be informed about the beliefs of the other side, and therefore they pay close attention to what they perceive to be dissonant news broadcasts. These findings also raise interesting questions with regard to source cues. We assume that when people employ source cues they identify a source they like and acquire and retain information from that source. In this instance it does appear that people are employing source cues to identify a source as consonant or dissonant, but once they have made this identification, they are better able to recall the information from the dissonant source. This finding supports the ideas presented in the theory of cross-cutting information (Mutz 2006; 2002; Mutz and Martin 2001), as individuals may well be turning to the media in order to have their political beliefs challenged perhaps because their beliefs are not often challenged in their daily lives. Also, it is quite possible that these respondents are simply projecting their beliefs onto these consonant media sources. Assume an individual holds certain political beliefs and thinks the way a specific news network reports stories is consistent with her beliefs. It is possible that that network can run a story on a specific topic that actually challenges her beliefs, but because it is run by a news network she purportedly agrees with, this person assumes that it must be consistent with her ideology and, when questioned, will assert that the report was consistent with her ideology even when it was not.

61 Although these results are highly provocative, they are by no means definitive, largely due to the study’s reliance upon an experimental approach. The most important follow-up would be one that assesses viewers out in the wild. These respondents apparently were put on guard by the dissonant sources, but it is possible that this was due to their participation in the experiment. Liberals who casually watch Fox News, or conservatives who casually watch CNN, are perhaps less likely to be on guard. In addition, work that determines whether respondents are acquiring and remembering more information from ideologically dissonant news sources because they are attempting to formulate counterarguments to the information presented to them, as well as whether people are projecting their ideological views onto the individual news networks, would be welcome additions to this area of research. It is this task that I turn to in the next chapter.

62 Table 4.1 Information Recall by Ideology and Network Attribution

Recall

Constant 1.900 (.056)**

CNN Content -.004 (.051)

Ideology .024 (.028)

CNN Attribution .128 (.067)*

FNC Attribution .166 (.067)**

Ideology X CNN Att .056 (.033)*

Ideology X FNC Att -.099 (.033)**

Ideology X CNN Cont. .005 (.026)

N =218 LR Chi2=44.70 Prob>Chi2 =.00 Pseudo R2=.045 *p<.10, **p<.05; standard errors in parentheses

63 Table 4.2 Correct Recall by Ideology and Network Attribution

Correct

Constant 1.196 (.078)**

CNN Content .024 (.064)

Ideology .013 (.038)

CNN Attribution .318 (.092)**

FNC Attribution .392 (.091)**

Ideology X CNN Att .136 (.045)**

Ideology X FNC Att -.194 (.045)**

Ideology X CNN Cont. .009 (.032)

N =218 LR Chi2=128.41 Prob>Chi2 =.00 Pseudo R2=..122 *p<.10, **p<.05; standard errors in parentheses

64 Table 4.3 Incorrect Recall by Ideology and Network Attribution

Incorrect

Constant 1.186 (.081)**

CNN Content -.036 (.084)

Ideology .039 (.041)

CNN Attribution -.127 (.079)*

FNC Attribution -.206 (.101)**

Ideology X CNN Att -.070 (.032)**

Ideology X FNC Att .075 (.035)**

Ideology X CNN Cont. -.007 (.041)

N =218 LR Chi2=19.48 Prob>Chi2 =00 Pseudo R2=.023 *p<.10, **p<.05; standard errors in parentheses

65 10

6 FNC N/A 4 CNN

0 Recall Correct Incorrect

Figure 4.1 Predicted Counts of Recall Attempts, Correct Recall Attempts, and Incorrect Recall Attempts of Strong Liberals by Network Attribution

66 10

6 FNC N/A 4 CNN

0 Recall Correct Incorrect

Figure 4.2 Predicted Counts of Recall Attempts, Correct Recall Attempts, and Incorrect Recall Attempts of Strong Conservatives by Network Attribution

CHAPTER FIVE

I HEAR WHAT YOU’RE SAYING, I JUST DON’T BELIEVE YOU: COUNTERARGUING AND DISSONANT MEDIA SOURCES

One of the most important topics for contemporary researchers of the relationship between the news media, the American public, and the dissemination of political information is the extent to which individuals learn from information presented on television news broadcasts. In the previous chapter I demonstrated that, somewhat unexpectedly, individuals are better able to acquire and retain information presented to them by a news network that they believe to be ideologically dissonant from their political beliefs. This finding raises the obvious question of why individuals are so successful at acquiring and retaining information from dissonant news sources. I argue that this takes place because individuals are formulating counterarguments to news content presented to them by a network that they believe to be ideologically dissonant from their political beliefs. This question is important to answer because it can provide much needed insight into the cognitive motivation behind why individuals process television news in the manner in which they do. This chapter examines the impact of the ideological consonance or dissonance of a television news source on an individual’s propensity to formulate counterarguments. I will begin with a review of the existing literature that has examined the influence of consonant and dissonant information sources. That will be followed by a discussion of the experimental design and measurement techniques utilized for the undertaking. I will then report the results of my hypothesis test, and conclude with a discussion of the theoretical and practical implications of these findings.

68 Theoretical Background This chapter builds upon the findings of the two previous chapters, which have demonstrated that viewers, particularly strong ideologues, perceive specific networks to be both ideologically consonant with, and dissonant from, their political position, and that strong ideologues are better able to acquire and retain news content presented to them by the dissonant media source. The argument advanced here to account for this, which is supported by the majority of previous research, is that respondents better remember news content presented to them by dissonant news networks because these respondents are likely formulating counterarguments to the news content. Respondent perceptions of source credibility exert a great deal of influence on the decision to counterargue. A credible source is defined as a source that possesses both correct information and a willingness to communicate that information in an unbiased manner (Hass 1981). Early work in the area of source credibility demonstrated that respondents would internalize, or adopt, the position or attitude advocated by the credible source but not the position or attitude advocated by the source perceived to be lacking in credibility (Kelman 1961). These early findings are supported by the work of Baron and Miller (1969), who demonstrated that subjects generated a significantly larger number of counterarguments to a message attributed to a low-credibility source than they did to a message attributed to a high-credibility source. Applying these findings to the question at hand, we would predict that strong liberals and strong conservatives would be more likely to formulate counterarguments to news content presented to them by Fox News Channel and CNN, respectively, because these ideological groups tend to view these networks as being ideologically biased against their political position and thus not credible. Other scholars have found that the level of commitment respondents demonstrate toward their political beliefs influences their propensity to counterargue. According to Brock (1967) and Haas (1972), individuals who are strongly committed to their political positions, as self-identified strong ideologues are commonly thought to be, are much less tolerant of opposing political viewpoints. As a result, these individuals will often take an argumentative stance, in the process mustering counterarguments in order to resist the persuasive intent or content of a message. The work of Eagly and Chaiken (1995), Johnson and Eagly (1989), and Zuwernick and Devine (1996) support these early

69 findings. In this case, we would expect that strong liberals and strong conservatives would be more apt to take argumentative stances toward what these groups believe to be ideologically dissonant news content presented by Fox News Channel and CNN, respectively, and to formulate counterarguments to news content presented by these networks. Research on message forewarning also supports the notion that respondents are more likely to formulate counterarguments to news content presented to them by a dissonant media source. Forewarning has typically taken the shape of either warning the message recipient of the content of the message or warning the message recipient of the persuasive intent of a message (Papageorgia 1968). Although some scholars dispute this assertion (Cooper and Jones 1970; Mills and Aronson 1965), a number of experimental studies have demonstrated that respondents who are forewarned of the forthcoming content of a message are significantly more likely to formulate counterarguments to resist a persuasive attempt (Allyn and Festinger 1961; Hass and Grady 1975; Petty and Cacioppo 1986). Much more agreement exists with regard to forewarning of the persuasive intent of messages. These examinations have demonstrated that respondents who were forewarned of the persuasive intent of a forthcoming message were very likely to formulate counterarguments to the information presented (Hass and Grady 1975; Fukada 1986; although Petty and Cacioppo (1986) demonstrated that counterarguments increased only for issues of high salience). Also, not surprisingly, Allyn and Festinger (1961) demonstrated that those who received warnings regarding both message content and persuasive intent were much more likely to formulate counterarguments to the message than were those who received neither warning. Although this concept refers to actually giving a warning to respondents before initiating a study, I argue it is relevant in this case because many respondents in today’s political culture essentially always possess these forewarnings regarding the media outlets under study. Many citizens already have preconceived notions regarding the content presented by these networks as well as the persuasive intent of their broadcasts. For instance, your average strong conservative most likely perceives CNN as pushing a liberal agenda, and the average strong liberal likely thinks Fox News is the mouthpiece of the conservative movement. As a result, because in this case strong conservatives and

70 strong liberals are “forewarned” regarding the nature and intent of the messages CNN and Fox News Channel present, these strong ideologues are most likely to counterargue news content presented by these news networks. The use of source cues also supports the notion that respondents should be more likely to counterargue news content presented by dissonant news sources (Carmines and Kuklinski 1990; Zaller 1992; Mondak 1993a, 1993b). Early work in this area dealt with respondent identification (Kelman 1961) with a source (also referred to as attractiveness by Osgood and Tannenbaum 1955 or similarity by McGuire 1969). Respondents used their level of identification with a source as a gauge for which viewpoints to accommodate and which to reject. Simply put, individuals were found to be less likely to counterargue information when they perceived it as coming from a source with viewpoints they thought were similar to the views they held. These sentiments were echoed by other scholars who found that individuals who held negative preconceived notions regarding the provider, or source, of information were more likely to counterargue the points the opposing side presented (Hass and Grady 1975; Kiesler and Kiesler 1964). Kuklinski and Hurley (1994) connected these with their finding regarding the propensity of the “messenger” to overwhelm the “message”. These authors found that respondents were more likely to oppose government programs proposed by individuals with whom the respondents did not agree politically. Because strong liberals and strong conservatives tend to have negative preconceived notions regarding the ideological predispositions of Fox News Channel and CNN, respectively, it is very likely that these groups will formulate counterarguments to information presented to them by these news networks. Researchers who have found that many citizens use the news media as a source of cross-cutting information would argue that acquiring information from news networks in order to formulate counterarguments is exactly how the system should work. In their daily lives individuals tend to choose politically homogenous discussion partners (Mutz 2006; Bennet, Fisher, and Resnick 1994; Ulbig and Funk 1999; although Huckfeldt et al. (2004) argue that some disagreement persists among political discussion partners). However, this political homogeneity does not appear to be as prevalent with regard to citizens’ choice of media sources, as they appear to be more willing to turn to a dissonant

71 news media source (Mutz 2002; Mutz and Martin 2001) to either challenge or reinforce (or perhaps challenge and reinforce) their political views. Therefore, it stands to reason that these individuals will be more likely to formulate counterarguments to news content provided by dissonant media sources as a result of their apparent desire to reinforce their pre-existing political beliefs. Finally, research by Chaiken et al. (1989) and Zimmerman and Chaiken (1998) that specifically focused on the processing of incongruent information has demonstrated that respondents acquire and retain more incongruent information than congruent information specifically in order to formulate counterarguments to it. This is supported by the work of Edwards and Smith (1996), as well as Eagly et al. (1999), who found that information that was not congenial evoked more thoughts, and most importantly more counterarguments, than did congenial information. Once again, if this theory holds we would expect that strong conservatives and strong liberals should acquire and retain more of the news content presented by CNN and Fox News Channel, respectively, because they are formulating counterarguments to that content. Hypothesis Due to the overwhelming consistency of the findings of previous scholars, I believe that respondents are more likely to formulate counterarguments to news content presented to them by what they believe to be an ideologically dissonant news source. As a result, the hypothesis for this examination is as follows:

Hypothesis 1: Respondents will formulate significantly more counterarguments to news content presented by a dissonant media source than they will to content presented by a consonant media source or a media source given no attribution.

I anticipate that strong liberals will formulate more counterarguments to news content attributed to Fox News Channel than they will to news content attributed to CNN or given no network attribution. Likewise, I anticipate that strong conservatives will formulate more counterarguments to news content attributed to CNN than they will to news content attributed to Fox News Channel or given no network attribution. Further, I anticipate that moderates will formulate a greater number of counterarguments to news

72 content attributed to both CNN and Fox News Channel than they will to news content given no network attribution. Methodology To examine this question, a multistage experimental design, which involved exposing respondents to consonant and dissonant news sources, was implemented. This study builds on previous approaches in that the content received by the experimental subjects is held constant, but the network attribution of the news content is manipulated. The key aspect of this design involved the creation of both the consonant and the dissonant news sources. This was done by creating replications of original Fox News Channel and CNN content, manipulating the network attributions, and showing the news stories to a group of experimental subjects. After viewing the videotapes the respondents were asked, among other things, to list any thoughts, feelings, or opinions about the stories presented. The dependent variable for this analysis is the number of counterarguments presented by respondents. Therefore, it is essential to have a clear definition of what constitutes a counterargument, as well as a sound method to measure the concept. For the purposes of this examination, a counterargument is strictly defined as “any statement given by a respondent that presented a rebuttal to the information presented on the news broadcast.” This is admittedly a strict definition, as a statement of simple disagreement, such as “I think the reporters are wrong”, wherein the respondent did not present some sort of rebuttal or counter to the information was not classified as a counterargument. In order to measure the concept, respondents were asked to detail “any thoughts, reactions, or feelings” about news stories with which they were presented (see Iyengar, Peters, and Kinder 1982). Their responses were coded in order to gauge how many different counterarguments each subject presented.27 The total counterarguments variable simply presents a tally of how many counterarguments the respondents put forth. This variable ranged from zero, which signified those who chose not to counterargue, to eight, which signified the greatest number of counterarguments presented by an individual respondent. With regard to the independent variables of interest, it was first necessary to account for the attribution of the stories. Two dichotomous variables were created to

27 Average inter-coder correlation across the stories analyzed was .82.

73 account for whether the stories were attributed to CNN or Fox News. CNN attribution was coded 0 if the stories are not attributed to CNN and 1 if they are. Likewise, Fox News Channel attribution was coded 0 if the stories are not attributed to Fox News and 1 if they are. Political ideology was initially measured on the traditional 1 to 7 scale; this variable was recoded from -3 to 3, with -3 signifying “strong liberal” and 3 signifying “strong conservative.” It was also necessary to control for the original source of the story to make sure that the content from one network did not incite a greater number of counterarguments than content that originated from the other network. This was accomplished via a dichotomous variable coded 0 if the original source of the content was Fox News Channel and 1 if the original source of the content was CNN. In addition, because the main contention of this piece is that ideologues are more likely to form counterarguments to a network with which they think that they disagree, variables were created that interacted ideology with network attribution. Finally, political ideology was interacted with the original source of content variable to ensure that the statistical model for this examination was properly specified. Results The model in Table 5.1 presents the results regarding the number of counterarguments formulated, which has a mean of 1.44 and a standard deviation of 1.76. This model tests whether respondents who are presented news content by a dissonant source are more apt to formulate counterarguments than those presented with news content by a consonant source. Because the dependent variables are counts, a Poisson regression was utilized to investigate these hypotheses.28 The findings presented in Table 5.1 support my expectations. As predicted, ideology conditions the influence of network attribution on the number of counterarguments formulated. The influence with regard to news content attributed to CNN is illustrated in Equation 1: EQ 1: Influence of CNN Attribution = [.766 + .275 (Ideology)] The significant coefficient of .766 operating on CNN Attribution is the conditional effect for ideological moderates. This positive coefficient indicates that

28 Tests for dispersion indicated that a Poisson regression was appropriate for this examination.

74 merely attributing news content to CNN inspires more counterarguments from this ideological group in comparison to news content that receives no network attribution. The significant interaction indicates that ideology conditions the influence of attribution. If you will recall, the coding of liberals ranged from -3 to -1. Therefore, as respondents become increasingly liberal, attributing news content to CNN has lesser effect, ultimately reaching a point at which strong liberals formulate no more counterarguments to news content attributed to CNN than they do to news content given no network attribution. In contrast, the coding of conservatives ranged from 1 to 3. As a result, as respondents become increasingly conservative, the influence of attributing news content to CNN is substantially greater, as strong conservatives formulate more counterarguments to news content attributed to CNN than does any other ideological group. The conditional relationship with regard to news content attributed to Fox News Channel is illustrated in Equation 2: EQ 2: Influence of Fox News Channel Attribution = [.683 - .327 (Ideology)] In this case, the significant coefficient of .683 operating on Fox News Channel attribution is the conditional effect for ideological moderates. This positive coefficient indicates that ideological moderates formulate more counterarguments to news content attributed to Fox News than they do to news content given no network attribution. Ideology once again conditions the influence of Fox News Channel attribution. As respondents become increasingly liberal, the influence of attributing news content to Fox News Channel reaches its zenith, as strong liberals are the most likely of all ideological groups to formulate counterarguments to news content attributed to Fox News. In contrast, for strong conservatives, attributing news content to Fox News Channel has little substantive effect on the number of counterarguments formulated in comparison to news content given no network attribution. The coefficient for the content variable is substantively small and statistically insignificant. This demonstrates that, among moderates, news content that originates from Fox News Channel does not inspire moderates to formulate more counterarguments than does content originating from CNN, and vice versa. In addition, the coefficient for the interaction between political ideology and original source of the news content is small and highly insignificant, which demonstrates that news content originating from Fox

75 News Channel also does not inspire the formulation of more counterarguments among ideologues than does content originating from CNN, and vice-versa. Poisson regression coefficients are informative in regard to the direction of the effect of independent variables and in regard to the statistical significance of these variables. However, they are difficult to interpret and provide little information regarding the substantive impact of each independent variable. Therefore, predicted counts were calculated in order to highlight the actual difference in the number of counterarguments formulated by strong liberals, ideological moderates, and strong conservatives. I will first examine the influence of network consonance and dissonance on the counterarguing efforts of strong liberals. There is no great difference in the number of counterarguments formulated when news content is either attributed to CNN or given no network attribution, as strong liberals are predicted to formulate 0.69 and 0.731 counterarguments, respectively. However, there is a large substantive effect when this same news content is attributed to Fox News, as under this circumstance strong liberals are predicted to formulate 3.85 counterarguments, which is more than five times the number of counterarguments formulated than when the news content is given no attribution or attributed to CNN. A similar pattern emerges with regard to strong conservatives. In this case, there is no great difference in the number of counterarguments formulated by strong conservatives when news content is either attributed to Fox News Channel (0.482 predicted counterarguments) or given no attribution (0.649 predicted counterarguments). However, there is a large substantive effect when this same news content is attributed to CNN, as under this circumstance strong conservatives are predicted to formulate 3.18 counterarguments, which is more than four times the number of counterarguments formulated when the news content is given no attribution and more than seven times the number of counterarguments formulated when the news content is attributed to Fox News. The Figure also highlights the influence of network attribution on the propensity of ideological moderates to counterargue. In this case, a Fox News Channel or CNN attribution indicates slight bias to this ideological group and causes them to generate more counterarguments (1.5 and 1.4, respectively) than when the same content is

76 presented yet attributed to no source (.68 counterarguments formulated). Therefore, when a news outlet is not considered to be ideologically dissonant with a viewer’s previously held political beliefs, no greater propensity to counterargue emerges. However, when the reputation of the media source is perceived to be ideologically dissonant from the viewer, the propensity of the respondent to counterargue increases dramatically. Conclusion and Implications Examining the influence of media source consonance and dissonance on respondent propensity to counterargue has allowed me to demonstrate that respondents who are exposed to an ideologically dissonant media source are more likely to formulate counterarguments to the news content presented than are those exposed to an ideologically consonant media source. By holding news content constant while providing ideologically dissonant and consonant network attributions, I was able to demonstrate that the perceived consonance or dissonance of a news source does matter with regard to respondent propensity to counterargue. Self-identified strong liberals and strong conservatives formulate significantly more counterarguments when watching Fox News Channel and CNN, respectively, than when watching the network that they believe to be more suited toward their ideological predispositions. In addition, moderate viewers were more likely to formulate counterarguments to both networks than when the news content was presented without network attribution. What are the potential implications of these findings? These results clearly support the notion that respondent perceptions of source credibility, especially with regard to the perceived willingness of a source to communicate information in an unbiased manner, exert a strong influence on the decision to counterargue (Hass 1961; Kelman 1961; Baron and Miller 1969). As illustrated in chapter two, strong liberals and strong conservatives have expressed doubts about the ability of Fox News Channel and CNN, respectively, to present unbiased news content, and political moderates do not think that either network produces unbiased news. As a result, strong liberals and strong conservatives have been demonstrated to be much more likely to formulate counterarguments to content presented by Fox News Channel and CNN, respectively, and political moderates have been shown to be much more likely to formulate

77 counterarguments to information attributed to either network as compared to news content given no network attribution. These results also reinforce the problem that a reliance on cognitive heuristics presents to viewers when they evaluate news content (Kuklinski and Hurley 1994; Tversky and Kahneman 1974; Lau and Redlawsk 1997; Mondak 1994). In this case, it was clearly shown that many respondents relied on cues transmitted from network attribution when evaluating the nature of the news content presented. As expected, strong conservatives put forth more counterarguments to news content attributed to, and originating from, CNN, and likewise strong liberals put forth more counterarguments to news content attributed to, and originating from, Fox News Channel. Surprisingly, however, the same pattern emerged for strong conservatives evaluating Fox News content attributed to CNN, and for strong liberals evaluating CNN content attributed to Fox News Channel. This illustrates that these respondents are not necessarily counterarguing ideologically biased information; rather, they are making judgments about the nature of the news content based on their preconceived notions regarding the political leanings of the source. Essentially, regardless of content, the messenger appears to be dictating the nature of the message to viewers. These results also support the belief that “forewarning” about the ideological predisposition of certain news network has become ingrained in society. These respondents were given no information regarding the nature of the news content presented by these networks or the persuasive intent of their broadcasts. Based on these results, however, it is clear that the average strong conservative believes that CNN advocates a liberal position, and the average strong liberal believes that Fox News Channel is the conservative news network. Therefore, it appears that these ideological groups are perpetually “forewarned” regarding the nature and intent of the messages that CNN and Fox News Channel present, respectively, and, as a result, some respondents are likely to counterargue any news content presented by these news networks.

78 Table 5.1 Total Number of Counterarguments by Ideology and Network Attribution

Constant -.460 (.174)*

CNN Content .087 (.115)

Ideology -.010 (.084)

CNN Attribution .766 (.198)*

FNC Attribution .683 (.203)*

Ideology X CNN Att. .275 (.094)*

Ideology X FNC Att. -.327 (.098)*

Ideology X CNN Cont. -.010 (.056)

N =218 LR Chi2=123.57 Prob>Chi2 =.00 Pseudo R2=.154 *p<.05; standard errors in parentheses

79 5

3 FNC N/A 2 CNN

0 Strong Liberals Moderates Strong Conservatives

Figure 5.1 Counteraguments Formulated in Response to Network Attribution by Ideology

80 CHAPTER SIX

THE MESSENGER OVERWHELMING THE MESSAGE: CONCLUDING REMARKS

One of the primary goals of this study was to uncover why, in the face of a lack of systematic scientific evidence, public opinion surveys have consistently demonstrated that a substantial percentage of Americans perceive an ideological bias within television news media. Because the television news media help fill the vital role of purveyor of political information to the American public, it was very important to examine two questions: first, why do individuals perceive ideological bias in the news media, and, second, does this perceived ideological bias influence how people process television news? I argued that the “CNN” and “Fox News Channel” network labels transmit ideological cues to the viewer, which influences both the manner in which individuals’ perceive the ideological leanings of these news networks, as well as how they processed the news content presented by these two networks. The results presented here have largely supported this initial argument. This final chapter will proceed in the following manner. I will begin with a summary of the experimental results presented in the substantive chapters of this study. That will be followed by a discussion of the theoretical and substantive significance of these results. I will conclude with recommendations for future avenues of research connected to this topic. Perceptions of Bias In chapter three, I argued that attaching the “CNN” and “Fox News Channel” labels to news stories transmitted an ideological signal to the viewer, which could lead the viewer to perceive ideological bias and, as a result, cause him or her to raise cognitive roadblocks that prevent proper processing of the information. My findings lent strong support to this argument. Irrespective of the network the news stories originate from, or the actual content of the stories presented, the CNN and Fox News Channel labels send a

81 clear signal to the viewer that the content of the reports is liberally and conservatively biased, respectively. Not surprisingly, these perceptions are most pronounced among ideologues who believe that their views are most at odds with those of a specific network. Perceptions of liberal bias in connection with news content attributed to CNN are most pronounced among strong conservatives, and perceptions of conservative bias in connection with news content attributed to Fox News Channel are most pronounced among strong liberals. I was also able to demonstrate that these perceptions of bias were not lost among more moderate viewers, as this ideological group also identified a liberal bias in news broadcasts attributed to CNN and a conservative bias in news broadcasts attributed to Fox News Channel. Information Acquisition In chapter four, I investigated whether viewers of news broadcasts were better able to acquire and retain information presented by news networks that were either consonant with or dissonant from their political beliefs. Given the results of previous research efforts, I anticipated that respondents would be better able to acquire and retain information presented by news networks that were ideologically consonant with their political beliefs. Quite surprisingly, the exact opposite was true. Respondents were actually better able to acquire and retain information presented to them by ideologically dissonant news sources, as opposed to ideologically consonant news sources. Liberal respondents made more factual recall attempts, more of those recall attempt were correct, and fewer of those attempts were incorrect when news content was attributed to Fox News Channel than when news content was attributed to CNN or given no network attribution. Conservative respondents made more factual recall attempts, more of those recall attempt were correct, and fewer of those attempts were incorrect when news content was attributed to CNN than when news content was attributed to Fox News Channel or given no network attribution. Political moderates, who find both networks to be at least somewhat dissonant from their political beliefs, made more factual recall attempts, more of those recall attempt were correct, and fewer of those attempts were incorrect when news content was attributed to either Fox News or CNN than when that news content was given no network attribution.

82 Counterarguing These surprising results led to the obvious follow-up question pursued in chapter five: why were viewers acquiring and retaining more information from news content presented by ideologically dissonant news sources than by ideologically consonant news sources? I argued that respondents acquired and retained more information from news content presented by ideologically dissonant networks because they were formulating counterarguments to contest the information with which they were presented. The findings in this chapter largely support this argument. In this instance, strong conservatives formulated significantly more counterarguments to news content attributed to CNN than they did to news content attributed to Fox News Channel or given no network attribution. Strong liberals, on the other hand, formulated significantly more counterarguments to news content attributed to Fox News Channel than they did to news content attributed to CNN or given no network attribution. Political moderates formulated significantly more counterarguments to news content attributed to both CNN and Fox News than they did to news content given no network attribution. Implications What are the larger implications of the results of this study? From a theoretical standpoint, these results lend credence to the use of cognitive heuristics, but also to the problems associated with their use (Carmines and Kuklinski 1990; Mondak 1993a, 1993b; 1994; Kuklinski and Hurley 1994; Tversky and Kahneman 1974 Lau and Redlawsk 1997). In this study the deployment of source cues clearly made a difference; respondents were exposed to virtually identical, ideologically neutral news content “messages”, yet their evaluations of the ideological nature of those messages varied significantly depending upon which news network happened to be the “messenger”. Clearly, the use of heuristics did not serve these individuals well, as it led them to incorrectly categorize the information. This is problematic because, in a real-world setting, this confusion regarding the ideological nature of information could result in confusion regarding the beliefs of political candidates, as well as confusion about the political ideals that liberals and conservatives strive to uphold. The results of this study also provide support for theories regarding cross-cutting information. Although a good deal of selectivity is taking place (Katz 1981; Klapper

83 1960; McGuire 1985; Axelrod 1973), a sizable number of individuals are exposing themselves to cross-cutting information (Mutz 2002; Mutz and Martin 2001). This exposure appears to be serving individuals fairly well, as they are learning more from the information, and engaging that information in a more in-depth manner, when news content is presented by a dissonant news source. However, individuals do not appear to be served as well by ideologically consonant news sources, as the content that they present has not been shown to inspire greater information acquisition, retention, or engagement. This can be a major problem considering the number of people who are drawn to ideologically consonant news sources. Perhaps in that instance the news is functioning as “background noise” rather than as political informant, which can present major difficulties with regard to informing the American public. In addition, these results provide great insight into the ability of news outlets both to inform the public and to serve as linkage institutions between the citizenry and the government. First, the fact that people acquire and counterargue information presented by dissonant news sources demonstrates that, at least on some level, the television news media are able to inform and engage individuals on a meaningful level. This indicates that the television news media have the ability to fulfill their role as political informant to the American public and, as well, to serve as an institution linking the activities of government officials and the desires of the citizens these officials represent. Secondly, given the fact that mere attribution to a network clouds how people evaluate news content, it is very likely that, absent a dramatic change in either the tone of the content presented or the marketing of these news outlets, there is little these networks can do to dispel the perception of themselves as ideologically biased. In chapter three I demonstrated that the news content presented was deemed neutral when given no network attribution, which indicates that the content these networks presented was not perceived to be promoting a biased view. However, merely attributing this news content, which was perceived to be ideologically neutral, to either Fox News Channel or CNN tremendously alters the manner in which the content is perceived. Because respondents perceived ideological bias in news content presented by these networks in a situation wherein the content itself has been deemed to be ideologically neutral, it seems that there

84 is little that these networks would be able to do in a real-world setting, outside of perhaps an unlikely to happen total network overhaul, to alter these perceptions. Finally, these results tell us a great deal about the nature of the typical American citizen. From an optimist’s perspective, these findings dispute the notion of the citizenry as being “woefully ignorant” (Converse 1964). As was illustrated in chapter three, absent a network attribution, respondents were able to correctly classify ideologically neutral information presented on news broadcasts as being ideologically neutral. The empirical findings presented in chapter four demonstrated that respondents, when adequately motivated by attribution of the news to a politically dissonant news provider, were able to acquire and correctly recall information presented to them on news broadcasts, and the findings in chapter five highlighted the propensity of those faced with politically dissonant news organizations to engage news content in a manner that was in-depth enough to allow them to formulate counterarguments to the information presented. When considered both individually and in their totality, these results strongly support the notion that, when cognitively motivated, citizens do indeed possess the capacity to actively engage the news in a meaningful way. This is good news, as an informed citizenry is a vital part of a healthy democracy. However, all the news is not good. Respondents were often cognitively lazy when evaluating content presented by news networks. When respondents were presented news content accompanied by a network attribution, the network attribution altered the manner in which respondents evaluated content. This network attribution allowed the respondents to engage in peripheral processing, which in this case led citizens to misperceive the ideological nature of the news content. In addition, respondents did not perform very well when asked to acquire and engage information presented to them by ideologically consonant networks. These findings do not bode well for the health of our political system, as many of those who flock to consonant news sources simply because those sources are consonant are likely to be less informed, and thus less effective, citizens. These results also indicate that the “myth” regarding the television news media being ideologically biased has become ingrained in our social reality. Perhaps, however, this is not the dire circumstance that many have predicted. The conventional wisdom has

85 been that perceptions of ideological bias lead to less effective information acquisition from the content presented. However, at least among these experimental subjects, perceptions of ideological bias actually resulted in greater levels of effectiveness in information acquisition. Because the experimental groups that received no network attribution indicated little difference in the ideological nature of the content presented by CNN and Fox News Channel, citizens as a whole would seem to be better off watching the network they believe to be ideologically dissonant because the news content with which they would be presented would not differ significantly, and the viewer would be better informed because they would be motivated to actively engage the content. Future Directions for Research Although the results of this study are highly provocative, they are by no means definitive, largely due to the study’s reliance upon a homogenous student sample. These respondents apparently perceived ideology bias and were put “on guard” by attribution of the news content to dissonant news sources, but it is possible that this was due to their participation in the experiment. Although I do not believe that the results presented in this work would change in any significant way, the most obvious follow-ups would be non-experimental studies, experimental studies that are unobtrusive, and studies that assess a sample of non-student respondents. The fact that these results support the notion that network attribution exerts a strong influence on people’s ideological perceptions of television news raises interesting questions regarding the strength of the ideological signal the network cue actually transmits. An interesting study would examine whether network attribution sends such a strong ideological signal that it can overshadow the cue sent by a biased news story. Specifically, is the ideological cue transmitted by CNN so strong that it would cause respondents to perceive a news story with an obvious conservative bias as having a liberal slant? Conversely, would attributing a liberally biased story to Fox News Channel enhance the extent to which respondents perceive it to be conservative? If this is indeed the case, these results could have interesting implications with regard to how the news media serves the American public. An additional extension of this line of research could include experimental research that explores people’s perceptions of CNN, Fox News Channel, and perhaps

86 even other news networks. Studies that explore people’s opinions of the networks and the personalities employed by the networks, as well as people’s perceptions of the veracity of claims of ideological bias made by liberal and conservative media watchdog groups, would be welcome additions to this field of research. Such studies would conceivably provide greater leverage on what factors influence perceptions of ideological bias, how permanent (or transient) these perception of ideological bias are, and what relationship exists between perceived ideological bias and perceived journalistic quality. In conclusion, does Fox News Channel have a conservative slant? Is CNN liberally biased? In one important sense, the answers do not really matter. So long as the viewers perceive ideological bias - and they do - allegations of media bias have the potential to produce very real consequences for American public opinion.

87 APPENDIX A NEWS STORY TRANSCRIPTS

TEXT OF ORIGINAL CNN STORIES Iraqi Fighting Topping our news this evening a U.S. Army scout helicopter was brought down by enemy fire Friday, killing one pilot, wounding the other, then a sneak attack by insurgents masquerading as news reporters, according to the U.S. military. According to military officials, five enemy personnel pulled up to the crash site driving black and dark blue Mercedes. They were wearing black press jackets with “press” clearly written in English. The enemy personnel fired upon U.S. forces with small arms and rocket-propelled grenades. No U.S. troops were hit, and later, four suspects were detained. It was a one-two- punch tactic that was also employed in an earlier attack on a U.S. convoy. A 5,000- gallon fuel truck was set ablaze by RPG and small-arms fire after, first, a roadside bomb stopped the convoy. Overall, the number of attacks against U.S. troops is down, from about 50 a day two months ago to about 20 a day now. But the enemies of the U.S. continue to refine their methods. One military source was quoted as saying that “We are seeing a small uptick in the capability of the enemy. They are getting a little more complex. And for what reason, we don't know. But they are getting a little more sophisticated of late.” There's no letup in the U.S. counterinsurgency operations. In the last 24 hours, the U.S. conducted more than 1,500 patrols, launched 28 offensive operations, and captured 88 anti-coalition suspects.

North Korean Inspection A door appears to be opening up in North Korea. According to South Korean officials, there's going to be a U.S. delegation that is going over and visit one of the nuclear sites there. What makes this significant is that it would be the first time that you

88 would have foreigners to actually visit that nuclear site since inspectors were kicked out about a year ago. But what is important to note as well is that U.S. government officials say, look, this is a private venture, that this is not something that is sanctioned by the government. However, they do acknowledge that there is some interest in what happens. Will they get to that country? Will they be allowed to that site? All those things they are going to be keeping a close eye on, because while U.S. officials say they are focused on the six-party talks involving other nations to get North Korea to disarm, they say those talks appear to have stalled.

Female Suicide Bomber It was bloody and shocking, a suicide bombing at the Erez Crossing between Gaza and Israel carried out by a young mother of two. A scene of devastation where the Palestinian suicide bomber struck. The attack on Israeli troops at Gaza's main crossing into Israel is a grim return to bloodshed. According to Brig. Gen. Shamni of the Isreali Defense Forces, “A woman suicide bomber came into the worker's pass in Erez industrial site, and exploded herself, killing four, three soldiers and one civilian, and wounding several other people.” Among the injured were Palestinian workers caught in the blast. Thousands pass through the Israeli security post every day to work in Israeli factories or gain access to Israeli services that impoverished Gaza simply doesn't offer. According to the Israeli Army, the suicide bomber, who was a woman, had said she needed a permit to get into Israel because she needed urgent medical attention. But, as she was waiting to be processed and searched by a female Israeli soldier, she dropped to her knees, burst into tears, and detonated. The suicide bomber has been identified as a 22-year-old from Gaza City, married with two children. Hamas says it's the first time they've used a woman to kill. The Al Aqsa Martyrs Brigade, which claims joint responsibility, has used female bombers before. Her attack, they say, was in response to Israeli military incursions and the construction of what Israel calls a security barrier in the West Bank.

89 An Israeli governmental spokesman characterized the attack of this morning as particularly shocking, because, as a gesture of goodwill, Israel allows Palestinian workers to come into Israel. And the Palestinian terrorist organization took this opportunity in order to kill as many people as possible.

For its part, the Palestinian Authority has called for a mutual cease-fire to make room for peace talks. But this latest bombing and the possible Israeli response may only deepen the mistrust.

Marriage Proposal The White House is also considering another major new initiative, this one to promote and to sustain marriage in this country. The initiative would be supported by $1.5 billion of taxpayer money. With a million of the two million yearly marriages ending in divorce court, love and marriage may go together like a horse and carriage. But happiness isn't assured. According to Courtney Knowles of the Equality in Marriage Institute: “We like to sometimes call the marriage aisle today the aisle of assumptions, because we think a lot of people are walking down it thinking they know their partner, thinking they understand what life is going to be like together, but not really wanting to kind of burst that bubble of warm, fuzzy feelings.” One of the more interesting things about marriage and divorce is that, for a nation so concerned about such things, we don't keep good records. There's no official national data on divorce collected. Budget cuts put an end to that in the mid-1990s. Among the things we do know, we are getting married at an older age now than we were in the 1950s. And we also know that divorce rates rose after World War II, when soldiers returned home, and they rose again in the 60s, with the advent of no-fault divorce laws. But perhaps the most shocking statistic is the one not about marriage. One-third of all children born today will be born out of wedlock and into a single- parent home. According to Theodora Ooms of the Center for Law and Social Policy: “A lot of the problems we see in society today that have to do with the fact that children are being

90 raised in single-parent or broken homes or homes where there is a lot of instability. Children are much poorer for that reason. They don't do as well in school. They are much more likely to have children out of wedlock themselves.” And that logic alone in the minds of many is enough for the government to get involved. Saddam Directives And finally tonight, did Saddam Hussein really want to keep al Qaeda at arm's length? According to U.S. officials, as well as a document found with Saddam Hussein when he was captured, Hussein warned supporters to be wary of cooperating with what were described as “foreign jihadists coming into Iraq to fight Americans”. Some in the Bush administration have contended there was close cooperation between Saddam's government and al Qaeda. Officials say the document appears to have been written after Saddam lost power.

91 TEXT OF ORIGINAL FOX NEWS CHANNEL STORIES Iraqi Attack In Iraq today US authorities define well-coordinated attacks as being carried out by “an enemy that does not respect any values.” Three targets were hit in Karbala with devastating results. Two coalition army bases were hit along with the town council building in Karbala. Attackers used four suicide car bombs as well as machine gun and mortar fire. The car bombers were shot before they could get all the way into the bases which limited the number of casualties. At least six soldiers were injured, four from Bulgaria and 2 from Thailand, and another two dozen were wounded. We’re told that several American soldiers received minor injuries. As in most of these cases the majority of those injured or killed were Iraqi civilians. More than a half dozen died and more than 80 were injured. US officials describe the operations performed by Iraqi and foreign fighters as low-intensity and being carried out by those trying to turn back the hands of time in Iraq.

Libyan Inspection Nuclear weapons inspectors from the United Nations arrived in Libya today. Inspectors traveled from Italy to Libya to begin their initial investigation into the extent of Libya’s efforts to build nuclear weapons. Libyan leader Momar Khadafi announced just last week that his country would scrap its weapons of mass destruction program. The International Atomic Energy Agency says this first trip will demonstrate just how cooperative Khadafi will be in dismantling Libya’s uranium enrichment facilities. The agency will also investigate a suspected connection between the nuclear weapons programs in Libya and Iran.

Mad Cow Fears It took just one cow with mad cow disease for the department of Agriculture to launch a massive investigation. Investigators are focusing on where that cow came from,

92 what it was fed, and what happened to its meat as some of it might have landed on store shelves. Investigators said today that they have made progress in finding out more about where this cow came from. Officials say they believe it was imported to the U.S. in August of 2001 from Alberta, Canada. It came in with 73 other cows. Now determining exactly where this cow came from is crucial to this investigation. What they want to know is what’s called the “birth herd”, which may lead them to other cows who ate the same feed and may also have been exposed to mad cow. Inspectors are also still pulling meat from the shelves that may have come from the infected cow. Parts may have ended up in Oregon, Washington State, California, and Nevada. This first ever mad cow case in the U.S. has agriculture officials considering new regulations. Most agree changes will be coming. One option is more testing for mad cow. Last year just over twenty thousand out of three million cows were tested, which is low compared to Western Europe who tested ten million and Japan who tested one million. Still, experts say the danger to consumers from the meat is almost nil. Mad cow effects the brain and nervous system presumably leaving the muscle safe to eat. That, however, is not recommended.

Illegal Alien Story A major question for President Bush is whether his immigration policy will put teeth into the current law. Right now a business that hires an illegal immigrant is subject to a ten thousand dollar fine. It sounds pretty tough but the reality is business has little to worry about. The General Accounting Office says that there are an estimated two hundred thousand business employing illegal immigrants. In 1992, 1,000 businesses were fined; by 2000 that number dropped to 13. Last year President Bush spent 791 million trying to keep illegal immigrants from crossing the border, on top of the money spent on fences and other technology. However, he spent only 20 million trying to catch those who did cross illegally.

93 There have been some high profile busts such as Wal-Mart and Tyson, but investigators say it is very difficult to prove these businesses knew they were hiring illegal help because as lawyers point out the documents are pretty good and businesses are often caught between a rock and a hard place as employers have a duty not to discriminate against those who do have lawful documents. Others say this is hogwash because companies know who works for them but need the cheap labor so they won’t seek these people out. They say it is almost a don’t ask, don’t tell. Immigration reform advocates say the “the technology is out there but we don’t use it because we choose not to know. Employers want to go on hiring these people with the excuse that “the documents looked good to me.” Truly it is part of the national ambivalence that says on one hand we need cheap labor but on the other hand these people are breaking the law and should be punished. But if the federal government was intent on rounding these people up they could go to any home improvement store in Southern California and do so. They don’t, so clearly the want is not there.

Saddam Directives U.S. officials confirm that among the documents obtained from Saddam Hussein at the time of his capture was a directive from Hussein to his loyal followers for them not to join forces with foreign Arab fighters which may be coming across the border to fight coalition forces in Iraq. This directive appears to be authentic according to officials. It was essentially a strict warning from Hussein to his loyalists not to get too close to Islamic Jihadists entering Iraq. With this document and other evidence officials have essentially determined Hussein believed foreign Arab fighters were eager for a holy war against the west while his Baath party was eager to return to power. Critics are saying this document shows how thin the relationship between Hussein and Al Qaeda really was. Officials stand by their earlier characterization of that relationship due to the meetings they have logged between the two groups. They do concede that the document appears to show Al Qaeda fighters were not the means by which Hussein wanted to fight the coalition.

94 U.S. Commanders have told Fox that they believe very few foreign fighters are behind recent attacks in Iraq and that this document supports this fact. The numbers vary but officials suspect only a few hundred foreigners are in Iraq fighting the coalition. Commanders think suicide bombers may be foreigners because Iraqi’s don’t want that job. The tie between the two groups as far as planning is still unclear this document does show that Hussein didn’t want radical Islamics running the show. Senior Defense Officials point out that this document may not have even made it out of Hussein’s hole; it just shows that he wanted some control over the regime’s attacks against the coalition.

95 APPENDIX B INFORMED CONSENT I am a graduate student under the direction of Professor Jeffery Mondak in the Department of Political Science at Florida State University. I am conducting a research study on the news media as part of a class project.

If you have any questions concerning the research study, please call me at (850) 576- 1757 or send me an email message at [email protected]. Additionally, if you would like to speak directly with Professor Mondak, he can be reached at (850) 644-7302 or by e-mail at [email protected].

If you have any questions about your rights as a participant in this research, or if you feel you have been placed at risk, you can contact the FSU Chair of the Human Subjects Committee, Institutional Review Board, through the Vice President for the Office of Research at (850) 644-8633.

I am requesting your participation, which will involve watching a video of a news broadcast and answering a brief questionnaire. Your participation in this study is voluntary. The questionnaire is anonymous. The results of the study may be published, but your name will not be known.

Your completion of the survey instrument indicates that you have read the consent form and decided to participate in this study.

Joel F. Turner, Jr. Graduate Student, Florida State University

96 APPENDIX C

HUMAN SUBJECTS COMMITTEE ACCEPTANCE LETTER

97 APPENDIX D

SURVEY

Please answer every question. Circle/fill in one answer per item. Thank you.

1. Race/ethnicity

African-American Caucasian Hispanic Other: ______

2. Sex: Female Male

3. Year of birth: ______

4. Generally speaking, would you consider yourself to be a Democrat, a Republican, or an independent? (circle one)

1 2 3 4 5 6 7 Strong Weak Weak Strong Democrat Democrat Democrat Independent Republican Republican Republican

5. And would you consider yourself to be liberal, conservative, or moderate?

1 2 3 4 5 6 7 Strong Weak Weak Strong Liberal Liberal Liberal Moderate Conservative Conservative Conservative

6. We are interested in your thoughts about Phil Streetman, the anchor who delivered the news broadcast you viewed. Based on your impressions please rate the anchor on the following characteristics by circling ONE of the numbers. a. appearance?

1 2 3 4 5 6 7 8 9 10 poor good b. tone of voice?

1 2 3 4 5 6 7 8 9 10 poor good c. level of knowledge?

1 2 3 4 5 6 7 8 9 10 poor good d. trustworthiness?

1 2 3 4 5 6 7 8 9 10 poor good

98 e. objectivity?

1 2 3 4 5 6 7 8 9 10 poor good f. how would you rate this anchor’s overall performance?

1 2 3 4 5 6 7 8 9 10 poor good

7. We are also interested in the different ways people get their news. On average, how many hours a week do you spend watching news programming on television? less than 5 6-10 11-15 16-20 more than 20

8. Which television news network do you watch most frequently?

CNN FoxNews CBS NBC ABC MSNBC PBS OTHER______

9. How many days each week do you read the newspaper? ______

10. Which newspaper do you most often read? ______

11. Do you use any online news sources? Yes No

12. If so, which one? ______

13. We are interested in your assessment of ideological fairness in how news stories are presented. We would like you to assess the fairness of the stories our anchor reported by ranking each story on a from 1 (liberally biased) to 10 (conservatively biased). a. The story about fighting in Iraq? (same question for both attributions)

0 1 2 3 4 5 6 7 8 9 10 liberal neutral conservative b. The story about nuclear inspections in Libya? (North Korea for CNN Attribution)

0 1 2 3 4 5 6 7 8 9 10 liberal neutral conservative c. The story about mad cow disease? (Marriage initiative for CNN Attribution)

0 1 2 3 4 5 6 7 8 9 10 liberal neutral conservative d. The story about illegal aliens? (female suicide bomber for CNN Attribution)

0 1 2 3 4 5 6 7 8 9 10 liberal neutral conservative

99 e. The story about Saddam’s directives? (same question for both attributions)

0 1 2 3 4 5 6 7 8 9 10 liberal neutral conservative

14. What network produced these stories?

CNN FoxNews CBS NBC ABC MSNBC PBS No Affiliation

15. Here are a few questions about government in Washington. First, who has the final responsibility to decide if a law is constitutional or not?

The president Congress The Supreme Court

15b. Whose responsibility is it to nominate judges to the Federal Courts?

The president Congress The Supreme Court

15c. Which one of the political parties is more conservative than the other at the national level?

Democrats Republicans

15d. What is the main duty of the U.S. Congress?

To write legislation To administer the To supervise states’ President’s policies governments

15e. Which party controls the House of Representatives?

Democrats Republicans

15f. Which party controls the United States Senate?

Democrats Republicans

16. We are also interested in your evaluations of possible ideological bias in the news. Based on your impressions please rate each network by circling ONE of the answers. a. ABC

0 1 2 3 4 5 6 7 8 9 10 liberal neutral conservative b. CBS

0 1 2 3 4 5 6 7 8 9 10 liberal neutral conservative

100 c. CNN

0 1 2 3 4 5 6 7 8 9 10 liberal neutral conservative d. Fox News

0 1 2 3 4 5 6 7 8 9 10 liberal neutral conservative e. NBC

0 1 2 3 4 5 6 7 8 9 10 liberal neutral conservative f. MSNBC

0 1 2 3 4 5 6 7 8 9 10 liberal neutral conservative

17. There were many specific pieces of information included in each of the stories you viewed. Can you recall five specific facts from the story on:

a. fighting in Iraq? (same for both network attributions)

101 b. mad cow disease? (female suicide bomber for CNN Attribution)

18. In the space below we would like you to please list any thoughts, feelings, or reactions you have with regard to the broadcast you just viewed:

102 19. We are also interested in the various traits of our respondents. For each set of adjectives presented please circle ONE that you believe best describes yourself.

0 1 2 3 4 5 6 7 8 9 10 neat sloppy

0 1 2 3 4 5 6 7 8 9 10 careless careful

0 1 2 3 4 5 6 7 8 9 10 irresponsible responsible

0 1 2 3 4 5 6 7 8 9 10 organized disorganized

0 1 2 3 4 5 6 7 8 9 10 insecure secure

0 1 2 3 4 5 6 7 8 9 10 calm tense

0 1 2 3 4 5 6 7 8 9 10 nervous relaxed

0 1 2 3 4 5 6 7 8 9 10 stable unstable

0 1 2 3 4 5 6 7 8 9 10 cold warm

0 1 2 3 4 5 6 7 8 9 10 kind unkind

0 1 2 3 4 5 6 7 8 9 10 agreeable disagreeable

0 1 2 3 4 5 6 7 8 9 10 unsympathetic sympathetic

0 1 2 3 4 5 6 7 8 9 10 shy bold

0 1 2 3 4 5 6 7 8 9 10 introverted extroverted

0 1 2 3 4 5 6 7 8 9 10 reserved outgoing

0 1 2 3 4 5 6 7 8 9 10 talkative quiet

0 1 2 3 4 5 6 7 8 9 10 shortsighted perceptive

103 0 1 2 3 4 5 6 7 8 9 10 efficient inefficient

0 1 2 3 4 5 6 7 8 9 10 self-assured unselfassured

0 1 2 3 4 5 6 7 8 9 10 intelligent unintelligent

0 1 2 3 4 5 6 7 8 9 10 confident unconfident

20. What do you think this study was about?

104 REFERENCES

Allyn, J. and L. Festinger. 1961. “The Effectiveness of Unanticipated Persuasive Communications.” Journal of Abnormal and Social Psychology, 62: 35-40.

Alterman, E. 2003. What Liberal Media? The Truth About Bias and The News. New York: Basic Books.

American Society of Newspaper Editors. 1998. Journalism Credibility Project Survey.

Arendt, H. 1968. “Truth and Politics.” In Between Past and Future: Eight Exercises in Political Thought, ed. Hannah Arednt. New York: Viking Press.

Asch, S. 1952. Social Psychology. New York: Prentice-Hall.

Axelrod, R. 1973. “Schema Theory: An Information Processing Model of Perception and Cognition.” American Political Science Review, 67: 1248-66.

Baron, R. and N. Miller. 1969. “Credibility, Distraction, and Counterargument in a Forewarning Situation.” Proceedings of the 77th Annual Convention of the American Psychological Association, 4: 411-12.

Bennett, S., B. Fisher, and D. Resnick. 1994. “Speaking of Politics in the United States: Who Talks to Whom, Why, and Why Not.” Paper presented at the Annual Meeting of the Midwest Political Science Association, Chicago, IL.

Bennett, W.L. and R. Entman. 2000. Mediated Politics: Communication in the Future of Democracy. Cambridge: Cambridge University Press.

Bower, R. 1985. The Changing Television Audience in America. New York: Columbia University Press.

Brock, T.C. 1967. “Communication Discrepancy and Intent to Persuade as Determinants of Counterargument Production.” Journal of Experimental Social Psychology, 3: 296-309.

Burtless, G. 1995. “The Case for Randomized Field Trials in Economic and Policy Research.” The Journal of Economic Perspectives, 2: 63-84.

Burtless, G. and L. Orr. 1986. “Are Classical Experiments Needed for Manpower Policy.” Journal of Human Resources, 21: 609-39.

Cacioppo, J., & R. Petty. 1979. “Effects of Message Repetition and Position on Cognitive Response, Recall, and Persuasion.” Journal of Personality and Social Psychology, 37: 97–109.

105 Carmines, E. and J. Kuklinski. 1990. “Incentives, Opportunities, and the Logic of Public Opinion in American Political Representation.” In Information and Democratic Processes, eds. J. Freejohn and J. Kuklinski. Urbana: University of Illinois Press.

Center for Media and Public Affairs. 2000. “Press Release, October 30, 2000.”

Chaiken, S. 1980. “Heuristic Versus Systematic Information Processing and the Use of Source Versus Message Cues in Persuasion.” Journal of Personality and Social Psychology, 39: 752-766.

------1987. The Heuristic Model of Persuasion. In Social Influence: The Ontario Symposium 5, eds. M.P. Zanna, J. M. Oslon, & C. P. Herman. Hillsdale, NJ: Lawrence Erlbaum.

Chaiken, S., A. Liberman, & A. Eagly. 1989. “Heuristic and Systematic Information Processing Within and Beyond the Persuasion Context.” In Unintended Thought, eds. J. Uleman & J. Bargh. New York: Guilford.

Chomsky, N. 1987. Chomsky Reader. Pantheon Books.

Converse, P. 1964. “The Nature of Belief Systems in Mass Publics.” In Ideology and Discontent, ed. D. Apter. New York: Free Press.

Cooper, J. and R.A. Jones. 1970. “Self-esteem and Consistency as Determinants of Anticipatory Opinion Change.” Journal of Personality and Social Psychology, 14: 312-20.

Dalton, R., P. Beck, and R. Huckfeldt. 1998. “Partisan Cues and the Media: Information Flows in the 1992 Election.” American Political Science Review, 92: 111-26.

Ditto, P., and D. Lopez. 1992. “Motivated skepticism: The Use of Differential Decision Criteria for Preferred and Nonpreferred Conclusions.” Journal of Personality and Social Psychology, 63: 568-584.

Ditto, P., J. Scepansky, G. Munro, A. Apanovitch, and L. Lockhart. 1998. “Motivated Sensitivity to Preference-Inconsistent Information.” Journal of Personality and Social Psychology, 75: 53-69.

Druckman, J. 2001. “Evaluating Framing Effects.” Journal of Economic Pyschology, 22: 91-101.

------. 2001b. “On the Limits of Framing Effects: Who Can Frame?” Journal of Politics, 63: 1041-1066.

106 ------. 2001c. “Using Credible Advice to Overcome Framing Effects.” Journal of Law, Economics, & Organization, 17: 62-82.

Eagly, A. and S. Chaiken. 1995. “Attitude Strength and Resistance to Attitude Change.” In Attitude Strength: Antecedents and Consequences, eds. R. Petty and J. Krosnick. Hillsdale, NJ: Lawrence Earlbaum.

Eagly, A., S. Chen, S. Chaiken, and K. Shaw-Barnes. 1999. “The Impact of Attitudes on Memory: An Affair To Remember.” Psychological Bulletin, 125: 64-89.

Edwards, A. 1941. “Political Frames of Reference as a Factor Influencing Recognition.” Journal of Abnormal Social Psychology, 36: 34-50.

Edwards, K. and E. Smith. 1996. “A Disconfirmation Bias in the Evaluation of Arguments”. Journal of Personality and Social Psychology, Vol. 71: 5-24.

Eldersveld, S. 1956. "Experimental Propaganda Techniques and Voting Behavior.” American Political Science Review 50: 154-65.

Entman, R. 1989. “How the Media Affect What People Think.” Journal of Politics. 51: 347-70.

Fairness and Accuracy In Reporting. 2003. “Spectrum Narrows Further in 2002: Progressive, Domestic Think Tanks See Drop.” In Extra, July/August.

Freedman, J. and D. Sears. 1967. “Selective Exposure.” In Advances in Experimental Social Psychology, Vol. 1, ed. L. Berkowitz. New York: Academic Press.

Frey, D. 1986. “Recent Research on Selective Exposure to Information.” In Advances in Experimental Social Psychology Vol. 19, ed. L. Berkowitz. New York: Academic Press.

Fukada H. 1986. “Psychological Processes Mediating the Persuasion Inhibiting Effect of Forewarning in Fear Arousing Communication.” Psychological Reports, 58: 87- 90.

Gallup Organization. 2003. “Are the News Media Too Liberal?” www.gallup.com/poll/releases/pr031008.asp.

Garfinkel, I., C. Manski and C. Michalopolous. 1992. “Micro Experiments and Macro Effects.” In Evaluating Welfare and Training Programs eds. C. Manski and I. Garfinkel. Cambridge: Harvard University Press.

Gastil, J. and J. Dillard. 1999. “Increasing Political Sophistication Through Public Deliberation.” Political Communication, 16: 3-23.

107 Gerber, A. and D. Green. “Reclaiming the Experimental Tradition in Political Science.” In Political Science: The State of the Discipline, 3rd Edition, eds. H. Milner and I. Katznelson. New York: W.W. Norton & Co.

Giner-Sorolla, R., and S. Chaiken. 1994. “The Causes of Hostile Media Judgments.” Journal of Experimental Social Psychology, 30: 165-180.

Gitlin, T. 2000. Inside Prime Time. Berkley: University of California Press.

Goldberg, B. 2001. Bias: A CBS Insider Exposes How The Media Distort the News. Washington, D.C.: Regnery Publishing.

Gosnell, H. 1927. Getting out the Vote: An Experiment in the Stimulation of Voting. Chicago: University of Chicago Press.

Graber, D. 1980. Mass Media and American Politics. Washington, D.C.: Congressional Quarterly Press.

Greenwald, A. and J. Sakumura. 1967. “Attitude and Selective Learning: Where are the Phenomena of Yesteryear?” Journal of Personality and Social Psychology, 7: 387-397.

Groeling, T. and S. Kernell. 1998. “Is Network News Coverage of the President Biased?” Journal of Politics, 60: 1063-87.

Groseclose, T. and J. Milyo. 2005. A Measure of Media Bias. Quarterly Journal of Economics, 120: 1191-1237.

Gunther, A., and C. Christen. 2002. “Projection or Persuasive Press? Contrary Effects of Personal Opinion and Perceived News Coverage on Estimates of Public Opinion.” Journal of Communication, 177-194.

Gunther, A., and K. Schmitt. 2004. “Mapping Boundaries of the Hostile Media Effect.” Journal of Communication, 54: 55-70.

Gussin, P. and M. Baum. 2004. “In the Eye of the Beholder: An Experimental Investigation Into the Foundations of the Hostile Media Phenomenon.” Paper presented at the 2004 annual meeting of the American Political Science Association, Chicago, IL.

Habermas, J. 1989. The Structural Transformation of the Public Sphere. Cambridge: MIT Press.

Hallin, D. 1992. “Sound Bite News: Television Coverage of Elections, 1968-1998.” Journal of Communication, 15: 5-24.

108 Hass, R. 1972. “Resisting Persuasion and Examining Message Content: The Effects of Source Credibility and Recipient Commitment on Counterargument Production.” Doctoral Dissertation, Duke University.

------1981. “Effects of Source Characteristics on Cognitive Responses and Persuasion.” In Cognitive Responses in Persuasion, eds. R. Petty, T. Ostrom, and T. Brock. Hillsdale, NJ: Lawrence Earlbaum.

Hass, R. and K. Grady. 1975. “Temporal Delay, Type of Forewarning and Resistance to Influence.” Journal of Experimental Social Psychology, 11: 459-69.

Hastorf, A. and H. Cantril. 1954. “They Saw a Game: A Case Study.” Journal of Abnormal and Social Psychology, 49: 129-134.

Hofstetter, R. 1976. Bias in the News: Network Television Coverage of the 1972 Election Campaign. Columbus: The Ohio State University Press.

Hovland, C. 1959. “Reconciling Conflicting Results Derived From Experimental and Survey Studies of Attitude Change.” American Psychologist, 14: 8-17.

Huckfeldt, R., P. Johnson, and J. Sprague. 2004. Political Disagreement: The Survival of Diverse Opinions Within Communication Networks. New York: Cambridge University Press.

Iyengar, S. 1991. Is Anyone Responsible? How Television Frames Political Issues. Chicago: University of Chicago Press.

Iyengar, S., and D. Kinder. 1987. News That Matters. Chicago: University of Chicago Press.

Iyengar, S., M. Peters and D. Kinder. 1982. “Experimental Demonstrations of the ‘Not- So-Minimal’ Consequences of Television News Programs.” American Political Science Review, 76: 848-58.

Jones, E. and J. Aneshansel. 1956. “The Learning and Utilization of Contravaluant Material.” Journal of Abnormal and Social Psychology, 53: 27–33

Johnson, B. and A. Eagly. 1989. “The Effects of Involvement on Persuasion: A Meta-Analysis.” Psychological Bulletin, 106: 290-314.

Joslyn, M., and S. Ceccoli. 1996. “Attentiveness to TV News and Opinion Change in the Fall 1992 Election.” Political Behavior, 18: 141-70.

Kaiser Family Foundation. 2001. “National Survey of the Role of Polls in Policymaking.”

109 Katz, E. 1968. “On Reopening the Question of Selectivity in Exposure to Mass Communications.” In Theories of Cognitive Consistency: A Sourcebook, eds. R. Abelson, E. Aronson, W. McGuire, T. Holcomb, M. Rosenberg, and P. Tannenbaum. Chicago: Rand-McNally.

Keisler, C.A. and S.B. Keisler. 1964. “Role of Forewarning in Persuasive Communications.” Journal of Abnormal and Social Psychology, 68: 547-49.

Kelman, H. “Process of Opinion Change.” Public Opinion Quarterly, 25: 57-58.

Kerlinger, F. 1986. Foundations of Behavioral Research. New York: Holt, Rinehart, and Winston.

Kinder, D. and D. Palfrey. 1993. “On Behalf of Experimental Political Science.” In Experimental Foundations of Political Science, eds. Donald Kinder and Thomas Palfrey. Ann Arbor: University of Michigan Press.

Klapper, J. 1960. The Effects of Mass Communication. Glencoe, IL: Free Press.

Kuklinski, J. and N. Hurley. 1994. “On Hearing and Interpreting Political Messages: A Cautionary Tale of Citizen Cue-Taking.” Journal of Politics, 56: 729-51.

Lane, R. 1962. Political Ideology: Why The Common Man Believes What He Does. New York: Free Press.

Lau, R. and D. Redlawsk. 1997. “Voting Correctly.” American Political Science Review, 91: 585-598.

Leighley, J. 2004. Mass Media and Politics: A Social Science Perspective. New York City: Houghton Mifflin.

Levine, J. and G. Murphy. 1943. “The Learning and Forgetting of Controversial Material.” Journal of Abnormal and Social Psychology, 38: 507–17.

Lodge, M., K. McGraw, and P. Stroh. 1989. “An Impression-Driven Model of Candidate Evaluation.” American Political Science Review, 83: 399-419.

Lodge, M., M. Steenbergen, and S. Brau. 1995. “The Responsive Voter: Campaign Information and the Dynamics of Candidate Evaluation.” American Political Science Review, 89: 309-26.

Lowry, N. and D. Johnson. 1981. “Effects of Controversy on Epistemic Curiosity, Achievement, and Attitudes.” Journal of Social Psychology, 115: 31-43.

110 Maitre, H. 1994. “The Tilt to the News: How American Journalism has Swerved From the Ideal of Objectivity.” In Taking Sides: Clashing Views on Controversial Political Issues, 9th ed., eds. George McKenna and Stanley Feingold Guilford. Connecticut: Dushkin Publishing Group.

McGuire, W. 1969. The Nature of Attitudes and Attitude Change.” In Handbook of Psychology, 2nd ed., eds. Gardner Lindzey and Elliot Aronson. New York: Knopf.

------1985. “Attitudes and Attitude Change.” In The Handbook of Social Psychology, Vol. 2, 3rd ed., eds. E. Aronson and G. Lindzey. New York: Random House.

Miller, J. and J. Krosnick. 2000. “News Media Impact on the Ingredients of Presidential Evaluations: Politically Knowledgeable Citizens Are Guided by a Trusted Source.” American Journal of Political Science, 44: 301-15.

Miller, M. 2002. The Bush Dyslexicon: Observations on a National Disorder. New York: W.W. Norton and Company.

Mills, J. and E. Aronson. 1965. “Opinion Change as a Function of the Communicator’s Attractiveness and Desire to Influence.” Journal of Personality and Social Psychology, 2: 152-59.

Mondak, J. 1993a. “Public Opinion and Heuristic Processing of Source Cues.” Political Behavior, 15: 167-87.

------1993b. “Source Cues and Policy Approval: The Cognitive Dynamics of Public Support for the Reagan Agenda.” American Journal of Political Science, 37: 186-212.

------1994. “Cognitive Heuristics, Heuristic Processing, and Efficiency in Political Decision Making.” In Research in Micropolitics, eds. Michael X. Delli Carpini, Leonie Huddy, and Robert Y. Shapiro. Greenwich, Conn.: JAI Press Inc.

Mutz, D. 2002. “Cross-Cutting Social Networks: Democratic Theory in Practice.” American Political Science Review, 96: 111-26.

------2006. Hearing the Other Side: Deliberative Versus Participatory Democracy. New York: Cambridge University Press.

Mutz, D., and P. Martin. 2001. “Facilitating Communication Across Lines of Political Difference: The Role of Mass Media.” American Political Science Review, 95: 97-114.

111 Orne, M. 1962. “On the Social Psychology of the Psychological Experiment: With Particular Reference to Demand Characteristics and Their Implications.” American Psychologist, 17: 776-83.

Osgood, P. and P. Tannenbaum. 1957. “The Principle of Congruity in the Production of Attitude Change.” Psychological Review, 62: 42-55.

Ottati, V. 1990. “Determinants of Political Judgments: The Joint Influence of Normative and Heuristic Rules of Inference.” Political Behavior, 12: 159-79.

Ottati, V., M. Fishbein, and S. Middlestadt. 1988. “Determinants of Voters’ Beliefs About the Candidates’ Stands on Issues: The Role of Evaluative Bias Heuristics and the Candidates’ Expressed Message.” Journal of Personality and Social Psychology, 55: 517-29.

Ottati, V. and R. Wyer, Jr. 1990. The Cognitive Mediators of Political Choice: Toward a Comprehensive Model of Political Information Processing. In Information and Democratic Processes, eds. J. Freejohn and J. Kuklinski. Urbana: University of Illinois Press.

Papageorgis, D. 1968. “Warning and Persuasion.” Psychological Bulletin, 70: 271-82.

Parenti, M. 1997. “The Myth of the Leftist Media.”

Patterson, T. 1994. Out of Order. New York: Random House.

Peffley, M., J. Glass, and J. Avery. 2001. “Public Perceptions of Bias in the News Media: Taking A Closer Look at the Hostile Media Phenomenon.” Paper presented at the Annual Meeting of the Midwest Political Science Association, at Chicago, Illinois.

Petty, R. and J. Cacioppo. 1986. The Elaboration Likelihood Model of Persuasion. In Advances in Experimental Social Psychology, ed. L. Berkowitz.

Pew Research Center. 1997. “Liberal Bias Hurts Credibility.” MediaWatch. April.

Program on International Policy Attitudes. 2003. “Misperceptions, the Media, and the Iraq War.” www.worldpublicopinion.org/pipa/articles/international_security _bt/102.php?nid=&id=&pnt=102&lb=brusc.

Project for Excellence in Journalism. 2000. “The Last Lap: Bush v. Gore.”

------2005. “The State of the News Media 2005.”

112 Riggle, E., V. Ottati, R. Wyer, Jr., J. Kuklinski, and N. Schwarz. 1992. Bases of Political Judgments: The Role of Stereotypic and Nonstereotypic Information. Political Behavior 14: 67-87.

Robinson, M. and M. Sheehan. 1983. Over the Wire and on TV. Russell Sage Foundation.

Robinson, M. and M. Clancey. 1985. Teflon Politics. Public Opinion 17: 14-18.

Rivlin, A. and T.M. Timpane. 1975. Ethical and Legal Issues of Social Experimentation. Washington, D.C.: Brookings Institution.

Smith, K., D. Johnson, and R. Johnson. 1981. “Can Conflict Be Constructive? Controversy Versus Concurrence Seeking in Learning Groups.” Journal of Social Psychology, 73: 651-53.

Sniderman, P., R. Brody, and P. Tetlock. 1991. Reasoning and Choice. Cambridge: Cambridge University Press.

Sweeney, K. and P. Gruber. 1984. “Selective Exposure: Voter Information Preferences and the Watergate Affair.” Journal of Personality and Social Psychology, 46: 1208-21.

Tversky, A. and D. Kahneman. 1974. Judgment Under Uncertainty: Heuristics and Biases. Science, 185: 1124-1131.

Ulbig, S. and C. Funk. 1999. “Conflict Avoidance and Political Participation.” Political Behavior, 21: 265-82.

Vallone, R., L. Ross, and M. Lepper. 1985. “The Hostile Media Phenomenon: Biased Perception and Perceptions of Media Bias in Coverage of the Beiruit Massacre.” Journal of Personality and Social Psychology, 45: 577-85.

Walsh, K. 2003. Talking about Politics: Informal Groups and Social Identity in American Life. Chicago: University of Chicago Press.

Weaver, P. 1972. “Is Television News Biased?” The Public Interest, 26: 57-74.

Woodward, B. 2002. Bush at War. New York: Simon and Schuster.

Zaller, J. 1992. The Nature and Origins of Mass Opinion. New York: Cambridge University Press.

Zillman, D. and J. Bryant. 1985. Selective Exposure to Communication. Hillsdale, NJ: Lawrence Erlbaum.

113 Zimmerman, J. and S. Chaiken. 1999. Attitude Structure, Processing Goods, and Selective Memory. Unpublished manuscript: New York University.

Zuwernick, J. and P. Devine. 1996. “Attitude Importance and Resistance to Persuasion: It’s Not Just Attitude Strength That Counts.” Journal of Personality and Social Psychology, 70: 931-44.

114 BIOGRAPHICAL SKETCH

EDUCATION

The Florida State University, Tallahassee, FL Ph.D., Political Science 2007 M.S., Political Science 2005

Georgia State University, Atlanta, GA 2002-2003 Masters Student, Political Science

The University of Georgia, Athens, GA 2002 Bachelor of Arts, Political Science Minor, Mass Communications

PUBLICATIONS

“The Messenger Overwhelming the Message: Ideological Source Cues and Perceptions of Bias in Television News”, forthcoming in Political Behavior

DISSERTATION

“The Messenger Overwhelming the Message: The Impact of Ideological Cues on Information Acquisition, Counterarguing, and Perceptions of Media Bias”

Dissertation Committee: Robert Jackson (chair), Charles Barrilleaux, Cherie Maestas, Jeff Mondak, Tim Salmon

WORKS IN PROGRESS

“Memorable Encounters: Ideology, Information Acquisition and Television News”

“I Hear What You’re Saying, I Just Don’t Believe You: Counterarguing and Dissonant Media Sources”

“Race and Attribution of Blame in the Aftermath of Hurricane Katrina” with Cherie Maestas and Lonna Atkeson

CONFERENCE PRESENTATIONS

“I Hear What You’re Saying, I Just Don’t Believe You: Counterarguing and Dissonant Media Sources”, presented at the annual meeting of the Western Political Science Association, Las Vegas, March 2007

115 “Memorable Encounters: Ideology, Information Acquisition and Television News”, presented at the annual meeting of the Midwest Political Science Association, Chicago, April 2006

“Race and Attribution of Blame in the Aftermath of Hurricane Katrina” with Cherie Maestas and Lonna Atkeson, presented at the annual meeting of the Southwest Political Science Association, San Antonio, April 2006

“The Messenger Overwhelming the Message: Ideological Source Cues and Perceptions of Bias in Television News”, presented at the annual meeting of the Southern Political Science Association, New Orleans, January 2005

“Dylan Glenn, Sanford Bishop, and the Race to Represent Georgia’s 2nd Congressional District”, presented at the annual meeting of the Southern Political Science Association, Savannah, November 2002

TEACHING EXPERIENCE

Instructor American National Government, The Florida State University • Summer 2005, Fall 2005, Spring 2006, Summer 2006

Media and Politics, The Florida State University • Summer 2006, Fall 2006, Spring 2007

Teaching Assistant Introduction to American Government, Georgia State University • Fall 2002

RESEARCH AND TEACHING INTERESTS

Media and Politics Electoral Behavior Public Opinion Political Psychology Southern Politics Campaigns and Elections Political Parties/Interest Groups Research Methods

HONORS AND AWARDS

Graduate Assistantship, The Florida State University 2003-present Graduate Assistantship, Georgia State University 2002-2003 Bascom Quillian Award for best graduate student paper, Georgia State University, 2003

116 PROFESSIONAL MEMBERSHIPS

American Political Science Association Midwest Political Science Association Southern Political Science Association Western Political Science Association

REFERENCES

Robert Jackson Associate Professor, The Florida State University 850-644-7307 [email protected]

Charles Barrilleaux LeRoy Collins Professor, The Florida State University 850-644-7643 [email protected]

Jeff Mondak James M. Benson Chair, The University of Illinois at Urbana-Champaign 217-244-6045 [email protected]

Cherie Maestas Assistant Professor, The Florida State University 850-644-7324 [email protected]

Stephen Nicholson Assistant Professor, The University of California-Merced 290-724-4335 [email protected]

117