Communist and Post-Communist Studies 46 (2013) 513–527

Contents lists available at ScienceDirect

Communist and Post-Communist Studies

journal homepage: www.elsevier .com/locate/postcomstud

Cold War redux in US–Russia relations? The effects of USmedia framing and public opinion of the 2008 Russia–Georgiawar

Oksan Bayulgen a, *, Ekim Arbatli b

a Department of Political Science, University of Connecticut, 365 Fairfield Way, Storrs, CT 06269-1024, USAb Faculty of Politics, NSU Higher School of Economics, Moscow, Russian Federation

a r t i c l e i n f o

Article history:Available online 26 September 2013

Keywords:US–Russia relationsRussia–Georgia warMedia framingPublic opinionCold War mentality

* Corresponding author.

0967-067X/$ – see front matter � 2013 The Regenthttp://dx.doi.org/10.1016/j.postcomstud.2013.08.003

a b s t r a c t

This paper examines the Cold War rhetoric in US–Russia relations by looking at the 2008Russia–Georgia war as a major breaking point. We investigate the links between media,public opinion and foreign policy. In our content analysis of the coverage in two majorUS newspapers, we find that the framing of the conflict was anti-Russia, especially in theinitial stages of the conflict. In addition, our survey results demonstrate that an increase inthe media exposure of US respondents increased the likelihood of blaming Russia exclu-sively in the conflict. This case study helps us understand how media can be powerful inconstructing a certain narrative of an international conflict, which can then affect publicperceptions of other countries. We believe that the negative framing of Russia in the USmedia has had important implications for the already-tenuous relations between the USand Russia by reviving and perpetuating the Cold War mentality for the public as well asfor foreign policymakers.� 2013 The Regents of the University of California. Published by Elsevier Ltd. All rights

reserved.

Even though the Cold War ended over twenty years ago, its legacy still permeates the discourse among politicians, themedia, and consequently the public at large in both Russia and the United States. Cold War remains the default paradigm forviewing US–Russia relations. One palpable example of this is the US Republican presidential candidate Mitt Romney’s pre-election comment in March 2012 that Russia is “without a question” the number one geopolitical foe for the United States(Oppel, 2012). Russian president Vladimir Putin also mirrored this Cold War mentality during his 2012 campaign with hisanti-American rhetoric (CBS News, 2012). More often than not, however, this discourse becomes rampant among elites andmedia amidst diplomatic crises and targets mainly the domestic audience.

There are certainly many historical, geostrategic, economic and political factors that shape the foreign policies of theUnited States and Russia towards one another. Nonetheless, the stereotypical negative image of the ‘other’ perpetuates thedistrust, suspicion and confrontation among foreign policy elites and limits the choices available for diplomatic engagement.What we are interested in this article is how this mentality of mutual bias is revived at certain times. We start with theassertion that the Cold War mentality is not a constant but that it is reignited during times of international crises, such as theKosovowar in 1999, Iraqi invasion of 2003, or the Ukrainian Orange Revolution of 2004. These breaking points in relations getreported and interpreted by the media in frames that are unfavorable and critical of the other side. The public generally lacksinterest in or access to information on international events, which by their very nature are very complex. As such, people take

s of the University of California. Published by Elsevier Ltd. All rights reserved.

O. Bayulgen, E. Arbatli / Communist and Post-Communist Studies 46 (2013) 513–527514

their cues from themedia in formulating their opinion of the other side. Themedia plays a key role in interpreting such eventsfor the broader public.

Meanwhile, the foreign policy elite uses these media frames and public perceptions to decide which foreign policybehavior is acceptable and justifiable. The media gives priority to certain issues and perspectives and puts them on thepolitical agenda of decision-makers. The connections between media framing, public opinion and foreign policy are certainlyfar from being simple and straight-forward. The literature, in fact, outlines a dizzying array of interactions between the public,leaders and the media in very complex and circular ways. For instance, the media is shown to have a strong agenda-settingfunction for foreign policy-makers (Denhan, 1997; Malek, 1996; Merrill, 1995; Robinson, 2000; Hallin, 1994). Alternatively, itis argued that the media is largely influenced by elite rhetoric (Entman and Page, 1994; Zaller and Chiu, 2000; Bennett et al.,2006) and that the public opinion relies on information shortcuts or heuristic cues (Sniderman et al., 1991; Popkin, 1994) thatare mainly opinions of trusted elites as reflected by the media (Auerbach and Bloch-Elkon, 2005; Groeling and Baum, 2008).Identifying the exact contours of this interdependent relationship and the direction of influence among these three centralvariables has been a core difficulty in research efforts; and in this paper our goal is not to resolve this complexity. Instead, wefocus on the particular role the media plays in constructing a narrative around such crises and shaping public opinion. Webelieve that themedia construction of crises and the consequent public opinion have had a significant effect on triggering andperpetuating the Cold War mentality that continue to influence relations between the two countries.

The 2008 Russia–Georgia war over the breakaway region of South Ossetia is one such breaking point in the relationsbetween the two countries. After a series of clashes between the South Ossetianmilitia and Georgian army troops throughoutthe summer, on the eve of August 7, Georgia launched a military attack on South Ossetia’s capital, Tskhinvali. The next day,Russia responded with a large-scale air and ground attack on Georgia. Not only did Russian troops gain control over SouthOssetian territory but they also entered Georgia proper and occupied Georgian cities of Poti and Gori quickly. On August 12,the two sides agreed to a French-brokered ceasefire and Russian president Medvedev ordered an end to Russian militaryoperations in Georgia. On August 26, the Russian government recognized the independence of South Ossetia and Abkhaziaand finally after much delay completed the withdrawal of majority of its troops from Georgia by October 8. The five-day warkilled hundreds, and left thousands of refugees in temporary shelters. The war was seen by many as the lowest point in US–Russia relations since the Cold War, if not an affirmation of the new Cold War itself (King, 2008; Sestanovich, 2008).

In this article, we focus on the way the US elite media framed this conflict and how this framing affected the way theAmerican public assigned blame in the war.1 We believe that the case study of the 2008 Russia–Georgia war offers a hard testfor the role the media plays in perpetuating the Cold War mentality. First of all, this was a war that the US was not directlyinvolved in. There were no vital US interests involved to give the media a mobilizing or “rally around the flag” mission.2

Therefore, there were not many reasons for the media to be unfavorable towards Russia in reporting the conflict.Secondly, given the complexity of the conflict, there was no obvious, undisputed aggressor in this war. There were in fact

two intertwined wars: an intrastate war between Georgia and its breakaway region South Ossetia and an inter-state warbetween Russia and Georgia. There were two contradictory interpretations of what actually happened and thus no over-whelming incentive for the US media to be favorable to Georgia over Russia. The Russian side (as well as the South Ossetianleaders) accused the Georgian government for starting the war. They claimed that Georgia attacked South Ossetia andcommitted acts of genocide against South Ossetians as well as Russian peacekeepers in the region. They argued that Russia gotinvolved only to protect the Russian peacekeepers and compatriots in South Ossetia. The Georgian side, on the other hand,argued that it wasmerely protecting its territorial integrity. The government claimed the attacks were a response to the heavyshelling of Georgian villages by South Ossetian militia and Russian troops. It accused Russia for using the skirmishes betweenSouth Ossetian rebel groups and the Georgian government as a pretext to invade Georgia and punish her for her pro-Westernorientation and democratic government.

Even though there was imbalance in the use of force by the militarily much stronger side, Russia, there were many third-party accounts from intelligence agencies, human rights organizations and scholars that challenged the view that Georgiawasvictimized by Russian aggressiveness (Chivers and Barry, 2008; Ertel et al., 2008; Hahn, 2008; Tsygankov and Tarver-Wahlquist, 2009; Tsygankov, 2010). For instance, the much-anticipated September 2009 report of the European Union’sIndependent International Fact-Finding Mission on the Conflict in Georgia (IIFFMCG) laid the blame for initiating the armedconflict at Georgia’s feet, but suggested that both parties share the blame for the long-simmering conflict (Lobjakas, 2009).Moreover, the European media outlets have used very different frames in their coverage of this conflict, demonstrating thatthe negative framing of Russia’s role in this conflict was not necessarily warranted by the events on the ground.3

Despite the lack of a compelling reason to take sides, the US media was accused of being biased against Russia in thecoverage of the conflict. One of the first people to openly voice this claimwas former Soviet president Mikhail Gorbachev. In aNew York Times opinion piece, dated August 20, Gorbachev accused the West of mounting “a propaganda attack against

1 A similar analysis of the Russian media coverage and public opinion responses no doubt would provide valuable and interesting comparisons but it isbeyond the scope of this article.

2 In this approach, the media tends to view their task in terms of reinforcing national consensus and is highly supportive of the government especiallywhen national interest is threatened.

3 ‘Campaign in the Air Waves: Europeans Divided on Georgian War: European Media’, EU-Russia Center (2008), available at http://www.eu-russiacentre.org. This article argues that the leading European media is divided onwhom to blame for the conflict, reflecting the political divisions within the EU as wellas the status of their relationship with Russia.

O. Bayulgen, E. Arbatli / Communist and Post-Communist Studies 46 (2013) 513–527 515

Russia, with the American news media leading the way” and he claimed that “the news coverage has been far from fair andbalanced, especially during the first days of the crisis” (Gorbachev, 2008). Along the same vein, the organization Fairness &Accuracy In Reporting (FAIR) argued that the US media seemed to use a Cold War mentality in presenting the conflict as a“black and white struggle between Moscow’s despotic aggression and Georgia’s pro-Western democracy” (FAIR, 2008).

Did the US elite media actually report and interpret the conflict in anti-Russia frames as some have claimed? If so, did theopinion of American public toward Russia becomemore negative after the conflict as a result of the media coverage? How didthe Cold War mentality that was reinforced in the media and reflected in public opinion affect the relations between Russiaand the United States in the aftermath of the conflict? These are the questions that our paper sets out to answer. The rest ofthe paper is organized as follows. In the next section, we overview the literature on framing and agenda setting functions ofthemedia to understand theways inwhich the USmediamay have contributed to the perpetuation of the ColdWarmentalityin the aftermath of the 2008 Russia–Georgia war. Next, we describe our research design and methods for data collection. Wefind that the two leading US newspapers, The New York Times and Wall Street Journal, were indeed more biased against theRussian side – at least initially – and that this type of negative media framing has had some effect on the way people assignedblame in this conflict. Finally, we discuss the implications of these findings, particularly in light of the already-sensitive andtenuous relations between the US and Russia.

1. Media framing and agenda setting

The role that the media plays in the construction of meaning through which people interpret the world is a well-studiedtopic in communications studies. Media is a powerful tool in telling us what issues to think about and how to think aboutthem. It is generally believed that the media has the power to influence public opinion by emphasizing certain issues overothers and advancing an agenda. The agenda-setting functions of the media may be especially strong when it comes to in-ternational news, because most individuals have no personal experience with and/or access to news events in foreigncountries. Thus, they may rely extensively on the media as the sole source of information and interpretation about the world.The agenda-setting role of the media in international news is also important for the way inwhich it may affect foreign policy.It can be argued that decision makers pay attention to the media coverage of international events in shaping their foreignpolicy decisions with the assumption that the degree of media attention to an issue reflects its importance to the public.

To understand the role of the media in contemporary life, researchers from a variety of disciplines draw on the concept offraming. According to Goffman (1974) a frame is a scheme of interpretation inwhich the particulars of events and activities towhich we attend are organized and made sensible. Gitlin (1980:7) defines media frames as “persistent patterns of cognition,interpretation, and presentation of selection, emphasis and exclusion, by which symbol-handlers routinely organizediscourse, whether verbal or visual.” Entman (1993) argues that frames “promote a particular problem definition, causalinterpretation, moral evaluation and treatment recommendation for the item described” (p.52).

There is a significant line of research that focuses onmedia frames, how they are created, and howdifferently they are usedby media outlets (Entman, 1991; Lee and Craig, 1992; Park, 2003). Among the framing studies, a few scholars have focused onone-sided framing in international news coverage, with the Palestinian–Israeli conflict getting the most attention (Viser,2003; Zelizer et al., 2002). But what exactly does one-sided media framing mean? Is there such a thing as ‘balanced framing’?

The major premise of framing theory is that an issue can be viewed from a variety of perspectives (Chong and Druckman,2007). Tuchman (1978) argues that news provide a frame that depicts a particular view of the world. If framing is the act ofconstructing reality, then the way an issue is framed reflects one view of the issue, out of many possibilities. For Reese (2001,10), for instance, framing is ultimately an “exercise of power”. Entman (1993, 55) argues that frames highlight some aspects ofreality while obscuring others, and as such they play a major role in the exertion of political power. It is through framing thatpolitical actors shape the texts that influence or prime the agendas and considerations that people think about (Entman, 2007,165).

Ultimately, then, no framing can provide a perfectly balanced presentation of news. This being said, however, in con-structing a reality for the audience, the media can sometimes frame the issue in a way that overwhelmingly favors one sideover the other in a current or potential dispute. Especially in international conflicts without clear victims and villains,journalists can legitimate the behavior of one side of the conflict by giving disproportionate salience to its justifications and byemphasizing certain images or words that delineate its victimization. Alternatively, they can project a condemnation of theother side by continually emphasizing its aggressive behavior and downplaying the specific explanations for its behavior.Through this kind of one-sided framing that persist over time and across media outlets, the media may systematically andinadvertently distribute power to particular groups or causes. Entman (2007, 166) defines this as content bias, referring to“consistent patterns in the framing of mediated communication that promote the influence of one side in conflicts”.4

4 It is important to note that one-sided framing or bias need not be conscious or intentional. In other words, it “could exist even where journalists andnews organizations possess no ideological goals, indeed where they consciously pursue balance” (Entman, 2007, 167). There are certain practical limitationsto media neutrality such as the inability of journalists to report all available stories and facts, and the requirement that selected facts be linked into acoherent narrative. Moreover, journalists may be stuck with the limited sources that are available to them, especially when covering international news. Forinstance, many point out to the regional and thematic imbalance that exists in news coverage of the world; that certain countries, regions and themes aremore consistently covered than others (Wilhoit and Weaver, 1983; Stevenson and Shaw, 1984; Hess, 1996; Chang, 1998; Wu, 1998; Golan and Wanta, 2003;Hatchen and Scotton, 2007).

O. Bayulgen, E. Arbatli / Communist and Post-Communist Studies 46 (2013) 513–527516

When the causes of war are not clear-cut as in this case and the identity of the villains and victims are open to debate, thetype of framing used in the media – even when there is no discernible intention – may make a significant difference inpeople’s construction of reality and perception of whom to blame. Therefore, our first goal is to understand how two leadingnational newspapers in the US framed the conflict and whether or not they overwhelmingly depicted Russia as an aggressor.

We are also concerned with the effects of media on public opinion, that is, the agenda setting function of the media. Theoriginal agenda-setting research proposed a moderate influence of media on public opinion. In their 1972 study, McCombsand Shaw made the important observation that media coverage of an issue increases the importance of that issue for thepublic. They also noted that the agenda-setting influence of massmedia is the transfer of salience from themedia to the publicagenda. This important media effect is particularly strong in the context of international news events. Since the public haslimited experience with foreign countries, it relies extensively on the media as an important source of information about theworld. For instance, many scholars found a strong correlation between exposure to foreign news and the salience of thoseissues for the public (Salwen and Matera, 1992; Wanta and Hu, 1993; Bissell, 2002; Soroka, 2003; Wanta et al., 2004). Somehave even associated exposure to foreign newswith relatively favorable feelings toward other countries (Reilly,1979;McNellyand Izcaray, 1986; Semetko et al., 1992).

Since the seminal work by McCombs and Shaw (1972), the agenda-setting hypothesis has been expanded beyond issuesalience to include attribute salience. Also known as the second level agenda setting, this theory states that media coveragenot only influences whatwe think about but also howwe think about that issue (Wanta et al., 2004). Frames influence publicopinion by “stressing specific values, facts and other considerations, endowing them with greater apparent relevance to theissue than theymight appear to have under an alternative frame” (Nelson et al., 1997, 569). Through framing, media generatescognitive and affective activities within individuals (McLeod et al., 1995). As Price and Tewksbury (1995, 382) put it, framingassumes that “the way in which choices are presented to people – the way choices are framed –will affect the likelihood thatpeople will select particular options”. This idea of ‘framing effects’ is what connects framing and agenda-setting theories,since both focus on people’s perception of reality through the meaning constructed by the media.

The second-level agenda setting theory has been predominantly used in mass communication research (McCombs et al.,1997; Golan and Wanta, 2001; Tedesco, 2001). However, the application of the theory in the area of international news stillremains limited. Golan andWanta (2003), for instance, examined the coverage of international elections on US newscasts andconcluded that negative coverage patterns led viewers to connect negative attributes to the perception of countries. Also,Wanta et al. (2004) found that the more negative coverage a nation received, the more likely respondents were to thinknegatively about that nation. Similar to these studies, using the agenda setting theory, we propose to study the relationshipbetween the media coverage of the August 2008 Russia–Georgia conflict and the US public opinion towards these countries.

2. Methods and research design

We employ two methodological strategies to answer the theoretical questions stated above. First, we employ a contentanalysis of two leading national newspapers in the US Second, we employ a national public opinion survey. Together, thesemethodological strategies provide for rigorous, yet broad-based, tests.

2.1. Content analysis

In the first part of the research, we use the content analysis method to study the media framing of the Russia–Georgiaconflict. Content analysis is appropriate here since it is extensively used to examinemedia coverage of events. It provides for asystematic analysis of detecting meaning, identifying intentions and describing trends in communication content (Holsti,1969; Neuendorf, 2002; Krippendorff, 2004; Riffe et al., 2005).

As for the choice of media, we focus on print media and quality press rather than popular press, because the former is themain source of foreign policy news with extensive and wide-ranging coverage (Beaudoin and Thorson, 2002; Paletz, 2002).Print media is also the main source for US foreign policy decision makers (Merrill, 1995; Malek, 1996).5 Even though it hasbeen shown that most people learn about the world fromwatching television (Larson, 1984), there is also extensive evidenceof an inter-media agenda setting process. Most local newspapers and television news programs take their cues on the salienceand attribute of issues from elite newspapers (Reese and Danielian, 1989; Protess and McCombs, 1991; Golan, 2006).

The newspaper articles analyzed in this study were drawn from the New York Times (NYT) and theWall Street Journal (WSJ).These newspapers were chosen because they are two of the leading media outlets in the United States with national cir-culation and are influential in setting the agenda for other print and visual media (Weiss, 1974; Gans, 1979; Merrill and Fisher,1980; Dreier, 1982). For instance, the NYT is considered the American “newspaper of record” (Chang et al., 1987).

We examined the news articles and opinion pieces that appeared in these newspapers between August 1 and September15, 2008. Even though the skirmishes between South Ossetian militia, Georgian and Russian troops began early on in the

5 Generally in the literature the distinction between popular and elite press focuses on the content, target audience, format and journalistic ethics. Elitepress is considered more serious, rational, text-based with a focus on ‘hard news’whereas popular press is seen as more sensational with more attention to‘soft news’ that is more colorfully people-oriented. It is also believed that elite press write for an educated, middle to upper class audience, who are morelikely to be the opinion leaders in society.

O. Bayulgen, E. Arbatli / Communist and Post-Communist Studies 46 (2013) 513–527 517

summer of 2008, the events were sporadic and thus were not getting comprehensive coverage in the US media. The end dateof September 15 was picked because beyond this date the conflict subsided considerably and media coverage became veryinfrequent.

The New York Times articles were drawn from the New York Times archive database and the Wall Street Journal articlesfrom the Factiva database. All these articles were searched with the key words “Russia” and “Georgia”. After eliminating thearticles that were not directly related to the recent conflict, a total of 240 articlesd97 from the New York Times and 143 fromthe Wall Street Journal – were selected for content analysis.

Upon a preliminary analysis of 10 randomly selected articles from each newspaper, we came up with 8 frames that wereoverwhelmingly used in the coverage of this conflict. Russian aggression frame was used to refer to the intensity and hostilitywith which Russian troops advanced into Georgia proper. Georgian aggression frame, on the other hand, referred to theGeorgian military offensive against the South Ossetians. Most articles also provided victimization frames for either side’sactions. The Georgian victimization frame referred to the suffering of the Georgian people in the hands of Russian troops. Forinstance, the following sentence was coded as using the Georgian victimization frame: “For more than aweek, the villages onthe roads running south from Tskhinvali, the capital of South Ossetia, some 20 miles to Gori, a central Georgian city nowunder Russian occupation, have been a corridor or grief, violence and crime” (Chivers, 2008).

Russian victimization frame predominantly covered the killings and suffering of the Russian troops and peacekeepingforces in the region. South Ossetian victimization frame was also present, referring to the suffering of South Ossetians as aresult of Georgian government actions and Georgian troops. In addition, the justification frames tried to legitimize someviolent actions by portraying them in an acceptable way. Russian justification frame covered topics like Georgia’s ambition tojoin NATO and the need to protect the lives and dignity of Russian citizens. The Georgian justification frame, on the other hand,focused on self-defense and territorial integrity claims by the Georgian government to explain its behavior. In addition, weincluded references to the Cold War period in the Cold War frame. The unit of analysis in coding for frames was the sentence.In the case that there was more than one frame that applied to a sentence, we picked themost dominant frame. In most cases,the dominant frame emphasized in a sentence was unmistakable, as in the case of the following sentence using the Russianaggression frame: “This brutality has been confirmed – although on a smaller scaledin the spectacle of the Russian armyoccupying a sovereign country, moving through it as it pleases, advancing and retreating at will, and casually destroying themilitary and civilian infrastructures of a young democracy as an astonished world watches” (Lévy, 2008).

We consider the combination of Russian aggression, Georgian justification and Georgian victimization frames to be an in-dicator for anti-Russia bias and alternatively Georgian aggression, Russian justification, Russian victimization and South Ossetianvictimization frames to be an indicator for anti-Georgia bias in the media. Therefore our first hypothesis in this paper is:

H1. There was an anti-Russia bias in the coverage of the Russia–Georgia war in the New York Times and Wall Street Journal.

2.2. Survey

For the second part of the research, we used data that was drawn from a survey conducted by the Center for SurveyResearch and Analysis (CSRA) at the University of Connecticut, using a Random digit dialing (RDD) telephone sample of adult,non-institutionalized residents of the contiguous United States. Telephone numbers were dialed daily from September 15,2008 through October 30, 2008, with interviewers attempting up to 6 calls to contact potential respondents.6 After securingcooperation, interviewers used Computer-Assisted Telephone Interviewing to administer questions to and record the re-sponses of 1000 participants. We asked two sets of questions in the survey. The first set included questions directly about theRussia–Georgia conflict and its coverage in themedia. We asked the respondents whether they have been following the newsof the conflict closely and which side they blamed in this conflict. In order to compare public opinion toward Russia beforeand after the conflict, we also asked a question about whether respondents think Russia is an American ally, wording it exactlythe same way as a CBS survey did in 2007.

Finally, the second set of questions in the survey was about how often the respondents get their news media from varioussources: local, network, and cable television, local and national newspapers, radio, and internet. The purpose for including thelatter set of questions was to cross-tabulate themwith the first set to see if there are any general patterns emerging betweenmedia coverage and public opinion. We also conducted regression tests to analyze the relationship between the public’sexposure to the news about the conflict and their propensity to blame one side over the other.

We assume that the negative framing of Russia as the aggressor has reignited the ColdWar image and fear of Russia amongthe US public. Based on second-level agenda setting theory, we also assume that there is a correlation between the level ofexposure to the news of the conflict and public opinion on Russia. Therefore, our next set of hypotheses is:

H2. US public opinion toward Russia has become more negative after this conflict.

H3. Those people who more closely followed the news of the conflict blamed Russia more than those who rarely followed or didnot follow the conflict.

6 The response rate was 31.4 percent. Although not as high as some national surveys such as the National Election Studies or the General Social Surveys,research has shown that lower response rates do not necessary yield lower quality data (Keeter et al., 2000).

Table 1Framing bias results.

Neutral Anti-Georgia Anti-Russia Mean coefficient

WSJ 31% (45) 5% (7) 64% (91) 0.18NYT 38% (37) 10% (10) 52% (50) 0.13Total 34% (82) 7% (17) 59% (141) 0.16a

Percentages are rounded up to the closest integer. Number of articles is in parentheses.a The t-statistic is 11.03 for the mean coefficient.Source: Authors’ database compiled from New York Times online archives and Factiva

O. Bayulgen, E. Arbatli / Communist and Post-Communist Studies 46 (2013) 513–527518

3. Content analysis results and discussion

As described above, the sample consists of 240 articles (143 fromWSJ and 97 fromNYT) for the period between August 1stand September 15th 2008. For both papers, the analysis includes both news stories and opinion pieces/editorials for the timeperiod covered. The articles were coded on several dimensions: framing in terms of aggression, justification and victimizationon both sides, and the type of sources used. Approximately 10% of the articles were double coded, using a random samplerepresentative of the percentage of total number of articles in each newspaper (10 articles fromNYTand 15 articles fromWSJ).

The intercoder reliability scores were calculated separately for frames and sources using three differentmethods to ensurevalidity: percentage, Scott’s pi and Cohen’s kappa. The scores for frames were 92% match for percentage and 0.87 for bothScott’s pi and Cohen’s kappa. For sources, they were 96% for percentage and 0.93 for both Scott’s pi and Cohen’s kappa. Thesescores are well over the conventional standards accepted in the literature (Lombard et al., 2002; Neuendorf, 2002).

The results for imbalance in framing were measured by forming a bias index that calculates mean coefficients. In thisindex, for each article the number of Russian aggression (Ragg), Georgian victimization (Gvict), and Georgian justification(Gjust) frames were added to arrive at the total number of anti-Russia frames (aRf). Similarly, for each article we addedGeorgian aggression (Gagg), Russian victimization (Rvict), Russian justification (Rjust), and South Ossetian victimization(SOvict) to arrive at the total number of anti-Georgia frames (aGf). Then, the difference of these totals was divided by the totalnumber of sentences (SeTot) in each article.

As it logically follows, the frame imbalance coefficient can be any value between 1 and�1: 1 representing an article that iscompletely anti-Russia and �1 representing one that is completely anti-Georgia in framing. In the most basic form of thisindex, only an article that scores 0 would be considered completely neutral. However, given how unlikely it is to receive ascore of precisely 0, we relaxed the assumption of neutrality by introducing a higher threshold: Neutrality is defined as anyarticle scoring between �0.10 and 0.10. Any score above this value was coded as anti-Russia and values below the thresholdwere coded as anti-Georgia.7

Our first hypothesis tests whether or not there is any anti-Russia bias in the reporting of the conflict. Our analysis confirmsH1, that negative framing of Russia exists in both newspapers. The results in Table 1 show that on total 59% of all the articlesused predominantly anti-Russia frames, as opposed to 34% neutral and only 7% anti-Georgia framing. The mean coefficientalso shows that both NYTandWSJ had on average anti-Russia framing of the conflict. A t-statistic of around 11.03 on themeanimbalance coefficient suggests that the mean 0.16 is significantly different from zero, indicating that anti-Russia framingexists.

It is interesting to observe some of the differences between the media coverage of NYT andWSJ. As seen in Table 1, 38% ofNYT articles, as opposed to 31% in WSJ, were neutral in terms of framing. The anti-Russia framing represented a bigger gap;WSJ having 64% (91 articles) anti-Russia frames compared to NYT’s 52% (50 articles). In terms of articles with an anti-Georgiabias, WSJ had half the amount compared to NYT, although both newspapers represented very low numbers of such articles (5and 10% respectively). In terms of mean bias coefficient for framing, WSJ was again confirmed to have more anti-Russia bias,although the difference is relatively low. When the bias index is broken down to the use of individual frames over time fornewspaper, we can also note other differences in the way the two newspapers reported the conflict. As seen in Table 2, thedominant frame in WSJ over time was Russian aggression, represented in 56% of all the frames coded, and the use of thisframe increased over time. Georgian victimization was the second dominant frame, although not nearly as prominent asaggression. NYT on the other hand, has preferred a more balanced use of Russian aggression and Georgian victimizationframes in its news coverage. This is an interesting finding in terms of the journalistic approach of the newspapers, suggestingthat while NYT has focused more on the humanitarian side of the conflict, WSJ has preferred to focus generally on powerpolitics.

The differences between the NYT and WSJ are further confirmed by the observation that WSJ had more Cold War framesthan NYT. NYT had 31 articles (32%) that directly refer to Cold War as opposed to the 52 articles (36%) in WSJ. Additionally,

7 To the best of our knowledge, there is no established measure in the literature that could be used as a threshold for neutrality. We acknowledge thatthis threshold is arbitrary, however, given our hypotheses that the media coverage will be one-sided, this threshold constitutes a harder test for our case bymaking the neutrality easier to achieve for a given article. We also account for the fact that this conflict was not one among equal powers, thus a slight anti-Russia framing imbalance could be expected. Additionally, to the best of our knowledge there is no established measure in the literature that could be usedas a threshold for neutrality.

Table 2Percentage of each frame in total number of frames over time.

Ragg Gjust Gvict Gagg Rjust Rvict Sovict

WSJAug 1–Aug 12 52% 9% 14% 6% 16% 0% 4%Aug 13–Aug 23 55% 6% 16% 3% 11% 1% 3%Aug 24–Sep 15 64% 3% 16% 4% 10% 0% 1%Mean 56% 6% 15% 4% 12% 1% 2%NYTAug 1–Aug 12 35% 8% 25% 7% 11% 1% 13%Aug 13–Aug 23 40% 3% 27% 9% 13% 2% 7%Aug 24–Sep 15 33% 6% 32% 7% 11% 0% 11%Mean 36% 5% 29% 8% 12% 1% 10%TotalAug 1–Aug 12 45% 9% 18% 6% 14% 0% 7%Aug 13–Aug 23 50% 5% 20% 5% 12% 1% 4%Aug 24–Sep 15 48% 5% 24% 6% 10% 0% 6%Mean 48% 6% 21% 5% 12% 1% 5%

Source: Authors’ database compiled from New York Times online archives and Factiva. Percentages are rounded up to the closest integer.

O. Bayulgen, E. Arbatli / Communist and Post-Communist Studies 46 (2013) 513–527 519

NYT had four times more Georgian aggression and five times more South Ossetian victimization frames, confirming that NYTdemonstrated less anti-Russia bias in its coverage. Another aspect to note in this breakdown is that both of the newspapershave refrained from using the Russian victimization frame. Although this could be attributed to the fact that Russian warcasualties were much less numerous, it may also indicate a conscious decision about who they believed the victim was.

The findings so far could raise the following question: Since both opinion pieces and news reports are included in theanalysis, can opinion pieces be driving the framing bias? In other words, are opinion pieces more negative against Russia thannews reports? A comparison of opinion pieces and news reports in these two newspapers shows that NYT was completelyneutral (a mean coefficient of 0.00) in its choice of opinion pieces and portrayed an anti-Russia bias in its news coverage. WSJ,on the other hand, produced an almost identical amount of anti-Russia bias both in its news coverage and the opinion pieces(with a difference of 0.01 between the two types – See Table 3). Thus the assumption that framing imbalance may be due toopinion pieces more than direct news analysis is not supported by the data.

The anti-Russia framing that we find in these newspapers can also be attributed to the sources that are used. News sourcesare significant for demonstrating who influences media content and how journalists tell a story or convey a certain message.Framing of an issue may depend on the source because different sources can provide different information to journalists andjournalists decide whether or not to use them extensively and howmuch salience to give them (Shoemaker and Reese, 1996).

The disproportionate use of sources by journalists may partly be a result of the ability of one side to convey its messagemore aggressively than the other side. For instance, in the Russia–Georgia conflict many have argued that Georgia won theimagewar by using aWestern public relations firm as well as a young, energetic, Western educated president to convince theworld that Russia is responsible for starting the conflict and committing the atrocities. As one of the New York Times articlesstated “it was not until four days after the conflict began – an eon in the 24-hour news universe – that a top Kremlin officialwas sent to CNN to counter Mr. Saakashvili” (Levy, 2008a).

In order to study this possible relationship between news sources and framing bias, we first coded for Russian andGeorgian sources separately. Any person and organization quoted or paraphrased providing information in stories was codedas a source. All the government, media and private sources fromRussiawere coded as Russian sources and those fromGeorgiaas Georgian sources. The unit of analysis in coding for the sources was the sentence. Next, we devised a measure of sourceimbalance index. In creating this, we used the same logic as the frame imbalance index. For each article, the number ofRussian sources (Rs) and Georgian sources (Gs) were coded and their difference was divided by the total number of sentences(SeTot) in the article. We applied the same threshold of neutrality to the source imbalance index as we did in frame imbalanceindex; any score between �0.10 and 0.10 represents a neutral article, a score above 0.10 means anti-Russia sourcing and ascore below �0.10 implies anti-Georgia sourcing.

Table 4 shows the distribution of the use of sources in the two newspapers. Both NYT and WSJ have relied almost equallyon Russian and Georgian sources in their news coverage and on total the use of sources were neutral in 62.5% of the articles.

Table 3Article types and framing bias.

News Opinion pieces Percentage of opinion pieces

WSJ 0.18 0.19 20% (29)NYT 0.15 0.00 11% (11)Total 0.16 0.14 17% (40)

Source: Authors’ database compiled from New York Times online archives and Factiva. Percentages are rounded up to the closest integer.Number of articles is in parentheses.

Table 4Source bias results.

Neutral Anti-Georgian Anti-Russian

WSJ 69% (98) 18% (26) 13% (19)NYT 53% (52) 23% (22) 24% (23)Total 62.5% (150) 20% (48) 17.5% (42)

Source: Authors’ database compiled from New York Times online archives and Factiva. Percentages are rounded up to the closestinteger. Number of articles is in parentheses.

Table 5Percentage and mean coefficient of bias in framing over time.

Neutral Anti-Georgia Anti-Russia Mean coefficient

WSJAug 1–Aug 12 19% (6) 10% (3) 71% (22) 0.21Aug 13–Aug 23 29% (22) 5% (4) 65% (49) 0.18Aug 24–Sept 15 46% (17) 0% (0) 54% (20) 0.14NYTAug 1–Aug 12 42% (8) 5% (1) 53% (10) 0.14Aug 13–Aug 23 34% (13) 8% (3) 58% (22) 0.13Aug 24–Sept 15 40% (16) 15% (6) 45% (18) 0.13TotalAug 1–Aug 12 28% (14) 8% (4) 64% (32) 0.18Aug 13–Aug 23 31% (35) 6% (7) 63% (71) 0.16Aug 24–Sept 15 43% (33) 8% (6) 49% (38) 0.14

Source: Authors’ database compiled from New York Times online archives and Factiva. Percentages are rounded up to the closest integer. Number of articlesis in parentheses.

O. Bayulgen, E. Arbatli / Communist and Post-Communist Studies 46 (2013) 513–527520

This finding is interesting given the imbalance in framing by both newspapers. However, when we looked at individual ar-ticles, we found a very high (0.567) and significant (p ¼ 0.00) association between the source and the framing imbalanceindices. In other words, an increase in the number of Georgian sources used in an article is correlatedwith an increase in anti-Russia framing. As it logically follows, the increase in the number of Russian sources would also mean an increase in anti-Georgia framing.

Although not tested in this study, the imbalance in framing may also be a result of the tendency of journalists to relyheavily on government officials to report on foreign events especially given the decline in international news bureaus inrecent years (Cohen, 1963; Sigal, 1973; Chang, 1993; Herman, 1993; Malek andWiegand, 1996). Cohen (1963, 28) for instanceargues that journalists are susceptible to the government’s claims because government officials are “prime sources of newsmerely by virtue of their positions in government.” Conformity to the government’s foreign policy agenda may indeed havebeen an important factor in the coverage of this conflict in the US media. The US government has taken an openly pro-Georgian stance during the course of the conflict, which is reflected in the media’s portrayal of events. Georgia was seenas a US ally aspiring to join NATO and as an example ofWestern-style democracy in the troubled region of Caucasus. The Tblisicables, part of the thousands of cables made available to news organizations by WikiLeaks, display the close relationshipbetween US diplomats and the Georgian government and the fact that Washington relied heavily on the Saakashvili gov-ernment’s accounts of what happened in the 2008 conflict (Chivers, 2010).

Finally, even thoughwe find anti-Russia framing in the coverage of this conflict, it is also important to analyzewhether thetype of framing varies depending on the stage of the conflict. Berry (1990) and Auerbach and Bloch-Elkon (2005) argue thatdifferent stages of international crises elicit different responses frommedia organizations and therefore framing of the issuesat hand vary dramatically from the onset phase of the conflict all the way to the de-escalation phase.8 In order to analyzewhether anti-Russia framing in the media is consistent throughout the coverage, we divided the coverage into 3 phases. Asthe first phase of the conflict (escalation stage), we took August 1 to August 12. August 1st was the official start of the war andAugust 12th was the ceasefire brokered by French President Sarkozy. The second phase (negotiation stage) took place be-tween August 13 and August 23. During this period, the official war was over but the terms of the ceasefire were notcompletely implemented, skirmishes continued between two sides and both sides were negotiating with each other as wellas with outside forces on the final status of the region. By August 23, Russia announced that it withdrew most of its troopsfrom Georgia. Finally, the third phase (de-escalation stage) took place during August 24 and September 15. There was nolonger any armed confrontation and both sides continued to make their case in international platforms.

8 In their analysis on the coverage of Bosnian crisis in the US media, Auerbach and Bloch-Elkon (2005) find a significant association between the positionof the press and the stage of crisis. For instance, they find that while the humanitarian meta-frame was predominant in the first two stages of the conflict,the domestic politics/economics meta-frame dominated at the de-escalation phase coverage. The authors also show that the newspapers were highlycritical of the Clinton administration’s policy towards Bosnia in the pre-crisis stage of the conflict, and they gradually shifted towards a more supportiveattitude as the crisis evolved through escalation and de-escalation stages.

Table 6Who do you think is more to blame for the conflict between Russia and Georgia–Russia or Georgia?Would you say that you have been following news of theconflict between Russia and Georgia very closely, fairly closely, not too closely, or not at all closely?

Russia Georgia Both Neither Don’t know Refused

Very closelyRow percentage 70% 18% 9% 2% 1% 0%Column percentage 11% 17% 17% 18% 0% 0%Somewhat closelyRow percentage 75% 10% 7% 1% 6% 1%Not too closelyRow percentage 63% 10% 3% 1% 21% 1%Column percentage 27% 27% 16% 31% 20% 41%Not at all too closelyRow percentage 24% 5% 2% 1% 68% 0%Column percentage 11% 15% 11% 28% 67% 6%Column percentage 0% 0% 2% 0% 4% 10%RefusedRow percentage 78% 0% 0% 0% 0% 22%Column percentage 1% 0% 0% 0% 0% 14%

Source: Authors’ calculations based on UCONN CSRA survey (Sept 15–Oct 30, 2008)

O. Bayulgen, E. Arbatli / Communist and Post-Communist Studies 46 (2013) 513–527 521

Given this breakup of the different stages of the conflict, we expected media to be quite neutral in the first stage of theconflict given the element of surprise, uncertainty about what is really happening, and a lack of clear direction from thegovernment.We further assumed that neutrality would disappear as propaganda from each side increased and the position ofthe US government became clearer.

The results in Table 5 suggest an opposite trend: that anti-Russia framing decreased as the conflict moved towards de-escalation. This was especially true in WSJ: anti-Russia framing steadily decreased as time passed, coupled with a decreasein anti-Georgia framing. This clearly shows that WSJ moved towards neutrality regarding both sides of the conflict over time.The mean bias coefficient also moves from 0.21 to 0.14, with a 33 percent decrease. Although there is no such linear pattern inNYT coverage, it is interesting to note that the number of anti-Georgia frames increased over time, and were tripled at the lastperiod compared to the onset phase. Overall, the anti-Russia frames slightly decreased as the conflict evolved.

There might be several reasons for this change. First of all, Russia as the winner in the conflict may have had more chancesof representation in the international arena than Georgia, thus over time had its point of view increasingly represented in themedia. By virtue of being the greater power, inevitably it has more leverage to dominate these platforms. Also, it may be thatthe US officials chose not to criticize Georgia until the conflict was over. Only at the de-escalation stage were the strategicblunders of the Georgian government voiced in the media. Finally, since military excursions were over at the later stages ofthe conflict, we spot less frames related to Russian aggression and Georgian victimization as time passes. Thus, the decrease inthe framing index can be attributed to the idiosyncrasies of the case, as well as to the fact that a conflict among unequalpowers will inevitably be reflected as such.

4. Public opinion survey results and discussion

The survey findings clearly show negative public opinion against Russia in this conflict. 57% of the people surveyed saidthat they blame Russia as opposed to only 9% that blame Georgia. Even though there were a significant number of “don’tknow” answers (27%), only 6% said that they either blamed both or neither. Similarly, in response to the question whetherthey feel that Russia is a close ally of the United States, 38% said that Russia is either a close ally or at least friendly but 51% saidthat Russia is either unfriendly or an enemy. This response is very similar to a question asked in a survey conducted by ABCNews/Washington Post between August 19 and 22, 2008.9 According to that survey, 39% of the respondents said Russia waseither a close ally or at least friendly, whereas 53% said she was unfriendly or an enemy.

In order to understand whether this conflict had any impact onwhat American people think of Russia, it would have beenideal to have a prior time series data on public attitudes on Russia. However since it is impossible to go back in time andconduct a survey of public attitudes towards Russia, we had to rely on a CBS News Poll of June 2007 (the only recent one to ourknowledge) as our baseline of public opinion. The ‘ally’ question from that survey that we repeated in our survey unfortu-nately does not sufficiently gauge the opinion on Russia but it is the closest we could find. According to the June 2007 CBS poll,68% of the respondents called Russia an ally or friendly (as opposed to our survey’s 38%) and 24% called Russia unfriendly or anenemy (compared to our 51%). These results – albeit preliminary – seem to lend some support to our second hypothesis thatthe US public opinion towards Russia has become more negative after the 2008 Russia–Georgia war.

Cross-tabulating the results between conflict and media questions demonstrates that there is an interesting pattern be-tween media consumption and public opinion. For instance, Table 6 shows that 61% of the respondents who blame Russia

9 The ABC News/Washington Post survey was based on telephone interviews with a national sample of 1108. Interviews were conducted by TNSIntersearch. Results were weighted to be representative of a national adult population.

O. Bayulgen, E. Arbatli / Communist and Post-Communist Studies 46 (2013) 513–527522

have followed the news very and somewhat closely as opposed to 38% that did not follow the news too much and not at all.Similarly, of all the people who followed the news of the conflict very closely, 70% blame Russia as opposed to 10% who blameGeorgia. The discrepancy in these percentages falls dramatically when looked among respondents who did not follow thenews at all: 24% blame Russia as opposed to 5%who blame Georgia. These results provide initial support to H3, that thosewhomore closely followed the news of the conflict blamed Russia more than those who rarely followed or did not follow theconflict.

Finally, in order to further test H3, we offer regression results from the survey. As the dependent variable we use thequestion “Who do you think is more to blame for the conflict between Russia and the Republic of Georgia?” All respondents whoanswered that “Russia”was to blame (n¼ 603) are coded as “1”; all respondents who answered “Republic of Georgia”, “both”,“neither”, or “don’t know” (n¼ 388) are coded as “0”; and all respondents who refused to answer (n¼ 9) are coded asmissing.This creates a dependent variable that focuses on whether the respondent has placed blame squarely on Russia or not. Toaccommodate the dichotomous structure of this dependent variable, we employ a logit estimator with the required weights.The key independent variable is derived from the question “Would you say that you have been following news of the conflictbetween Russia and the Republic of Georgia very closely, somewhat closely.?” It is labeled “Followed the Crisis” in the re-gressions. The response options are “not at all closely” (coded as 1, n¼ 196), “not too closely” (coded as 2, n¼ 257), “somewhatclosely” (coded 3, n ¼ 424), and “very closely” (coded as 4, n ¼ 107). Respondents who replied “don’t know” or refused toanswer are coded as missing (n ¼ 16). As such, the higher the value for the variable, the more the respondent claims to havefollowed the crisis. We expect it to be positively correlated with the dependent variable. The more that one is exposed to thenews that portrayed Russia as the aggressor, the more that she or he is likely to blame Russia for the conflict.

We also employ 10 control variables that range from other media behaviors and attitudes, overall attitudes towards Russia,as well as a number of typical demographic variables. First, the survey asks the respondents a series of 7 questions on theirnews media usage in 7 different media venues (for example, Local TV and National Newspapers). The four responses varyfrom “never” (coded as 1) to “everyday” (coded as 4). The 7 questions are aggregated to form “Overall Media Usage”. Thisvariable controls how much the respondent is generally exposed to news.

In addition, we control for the respondents’ self-reported level of “trust and confidence in the news media.” This is codedas 1 through 4 with higher values indicating a higher level of trust. Third, we control for respondents who may have a biasagainst Russia generally. Respondents are asked if they think that “Russia is a close ally.?” The four responses range from“unfriendly and an enemy” (coded 1, n ¼ 148) to “close ally” (coded 4, n ¼ 20). Although the Russia–Georgian conflict mayinfluence the respondent’s choice, the question is worded broadly and specifically asks about its relationship to the UnitedStatesdand the conflict, at most, only marginally affected the United States. As such, we are confident that this questionlargely taps into the respondents pre-existing opinion of Russia10; the higher the values, the more the respondent believedthat Russia is an ally of the United States.

The remaining 7 variables are typical controls inmost studies of political behavior. In concern for space, wewill omit manyof the question and response details. All of the questions and responses, though, contain standard wordings. These includedemographic controls, such as the Age of the respondent, whether the respondent is self-identified as beingWhite or not, andwhether the respondent is Male or not. We also control for the respondent’s self-placement on a five-point ideological scaleranging from “very liberal” (coded as 1, n ¼ 115) to “very conservative” (coded as 5, n ¼ 146) and on a five-point religiosity,which is measured as how frequently the respondent reports attending religious servicesdranging from “never” (coded as 1,n ¼ 114) to “every week” (coded as 5, n ¼ 322). Finally, we also code for the respondents education level, which varies from“grade school or less (0–8)” (coded as 1, n ¼ 7) to “post graduate (more than 4 years)” (coded as 6, n ¼ 218) as well as therespondent’s family income levels, which varies from “under $25,000” (coded as 1, n ¼ 103) to “$100,000 or more” (coded as5, n ¼ 209).

Table 7 reports the coefficients from the logit estimations. We estimate five models. The first, labeled “All Respondents”, isregressed on all respondents for which there is data. Among the 11 variables, only the key independent variable, Followed theCrisis, is statistically significant. As expected, it is positively correlated with the probability that the respondent blames Russiafor the crisis. Many Americans probably had little prior knowledge of Russia–Georgian relations and the ensuing politicalturmoil; it is, thus, likely that respondent’s attitudes towards the crisis have been shaped largely by the domestic newscoverage. As we have shown, this news coveragedat least in the elite outlets that we surveyeddhad typically framed Russiaas the aggressor. As such, we expected in H3 that the more an individual has followed the crisis, the more likely he or shewould blame Russia for the conflict. In fact, if all of the other variables are held at their means, respondents who followed thecrisis “very closely” are 49 percent more likely to blame Russia than respondents who followed the crisis “not at all closely.”

The absence of statistical significance for any of the other regressors speaks to the likelihood that respondents probablyhad no to little prior formed opinions about Russia and Georgia. Interestingly, opinions on whether Russia is to blame or notfor the crisis also withstand any prior attitudes about whether Russia is an ally or the respondent’s overall media usage.

To further test this finding, we analyze subgroups from the sample. The survey asked respondents “overall, where wouldyou say that you get most of your news from?” The response options include television (n ¼ 546), newspapers (n ¼ 133), theradio (n ¼ 109), the internet (n ¼ 158), family/friends (n ¼ 24), combination (n ¼ 16), don’t know (n ¼ 13), and refused toanswer (n¼ 1). This provided the opportunity to examine attitudes about crisis among thosewho use themain-stream-media

10 The regression results are substantively unchanged if this variable is omitted.

Table 7Blaming Russia for conflict.

All respondents Most news from TV Most news from newspapers Most news from radio Most news from internet

Followed the crisis 0.840*** 0.943*** 1.558*** 1.537*** 0.346(0.154) (0.188) (0.381) (0.437) (0.336)

Overall media usage 0.083 0.064 0.112 �0.643 0.060(0.160) (0.223) (0.440) (0.437) (0.434)

Trust media �0.021 �0.053 0.212** 0.029 �0.205**(0.036) (0.053) (0.097) (0.144) (0.084)

Russia as Ally 0.005 �0.193 1.119*** 0.032 �0.191(0.148) (0.199) (0.366) (0.402) (0.437)

Age �0.003 0.002 �0.023 0.027 0.038(0.008) (0.010) (0.024) (0.033) (0.031)

White 0.131 0.213 2.342** �0.763 �2.499**(0.311) (0.407) (0.975) (1.056) (1.184)

Male 0.061 0.401 �1.441** �1.717** 0.602(0.229) (0.341) (0.705) (0.872) (0.599)

Conservative 0.081 0.092 0.376* 0.149 �0.059(0.090) (0.123) (0.212) (0.306) (0.273)

Religiosity �0.033 0.024 �0.298 �0.239 0.007(0.077) (0.108) (0.227) (0.243) (0.221)

Education 0.125 0.267* 0.606* �0.546 �0.186(0.104) (0.156) (0.324) (0.392) (0.328)

Family income 0.001 �0.084 �0.595** 0.271 0.964***(0.097) (0.122) (0.251) (0.317) (0.253)

Constant �1.786* �1.711 �11.39*** 0.242 2.231(0.962) (1.330) (3.294) (3.161) (2.813)

Observations 635 620 622 621 622Subgroup observations 358 81 63 107F statistic 3.55*** 3.31*** 2.67** 1.67* 2.21**

Notes: Untransformed coefficients reported; Standard errors in parentheses.***p < 0.01, **p < 0.05, *p < 0.1.Source: Authors’ calculations based on UCONN CSRA survey (Sept 15–Oct 30, 2008)

O. Bayulgen, E. Arbatli / Communist and Post-Communist Studies 46 (2013) 513–527 523

(MSM) and those who do not. We assume that those who receive most of their news through the MSM (defined here astelevision, newspapers, and the radio) are likely to be exposed to similar framing.11 Although the Internet certainly offersnews from the New York Times, the Wall Street Journal, and others, there is also a much greater diversity of news sources,including first-person accounts. As such, we expect that respondents who have followed the crisis more closely and reportthat they receive most of their news from one of the three MSM sources are more likely to blame Russia for the crisis. Theremaining fourmodels support our hypothesis. The coefficient estimates for Followed the Crisis variable in themodels labeled“Most News from TV”, “Most News from Newspapers,” and “Most News from Radio” are all statistically significant andpositively related to the probability that the respondent blamed Russia for the crisis.12 This suggests that the more the re-spondents followed the crisis and reported that they receive most of their news from one of the three MSM sources, the morelikely that they were to view Russia as the aggressor in the conflict. Alternatively, in the model labeled “Most News form theInternet”, the key independent variable, Followed the Crisis, is far from statistically significant.

Although we cannot be sure of why this is, it is plausible that the diversity of news and commentary that is offered acrossthe internet has disrupted the pattern that has otherwise been consistently founddthat is, Russia was framed as theaggressor in the Russia–Georgia conflict by the elite American newspapers. As Chong and Druckman (2007, 119) put it“competitionmakes accessible a more representative sampling of underlying considerations than one-sided communicationsand therefore provides a broader frame of reference for locating one’s preference on an issue”.

5. Cold War mentality, public opinion and US–Russia relations

Each time when there is a crisis between Russia and the US, experts warn about the resurgence of a Cold War mentalitythat permeates the rhetoric and behavior of foreign policy elites. Despite the fact that the Cold War has been over for morethan two decades, the elites as well as the masses on both sides occasionally fall back on their preconceived notions andbiases and cannot seem to shake away the distrust and suspicion of the other side that characterized the relations throughoutmuch of the 20th century. In this paper, we set out to explain the role of the media in reviving and transmitting this mentalityby selectively using certain frames over others in the coverage of crises. Through a content analysis of the two leading elitenewspapers, we find that the US media did indeed provide an anti-Russia framing in the coverage of the war between Russia

11 As discussed earlier, this is referred to as the “inter-media agenda setting.”12 The F-Statistic for the “Most News from Radio”model is statistically significant, but only at the p < 0.1 level. Even though Followed the Crisis has strongstatistical significance, we need to be cautious about whether the entire model is distinguishable from the null hypothesis in which all of the coefficients arezero.

Fig. 1. Source: Gallup (1993, 1994, 2995, 1996, 2997, 1999, 2000, April 2001, March 2003, April 2003, 2006), Harris (October 2001, 2004, 2005) and CBS (2007)polls. The 2008 data comes from our own poll. Question: For each of the following countries, please say whether you consider it an ally of the United States,friendly, but not an ally, unfriendly, or an enemy of the United States. How about – [RUSSIA]?

O. Bayulgen, E. Arbatli / Communist and Post-Communist Studies 46 (2013) 513–527524

and Georgia in 2008. We find that the framing bias was not driven by opinion pieces or the sources that were used and that itvaried depending on the stage of the conflict, meaning that anti-Russia framing decreased over time as the conflict movedtoward de-escalation.

Our public opinion survey also provided us with the opportunity to test the agenda-setting role of the media. Thedescriptive statistics reveal that US public opinion of Russia became more negative after the 2008 crisis and that those whomore closely followed the news of the conflict blamed Russia more than those who rarely followed or did not follow theconflict. Our regression analysis further confirms that the more the respondents followed the crisis, the more likely they wereto view Russia as the aggressor in the conflict. Although with varying effects depending on the type of media outlet, the USmedia coverage has had an important role in perpetuating the Cold War mentality that Russia is an aggressive power thatshould not be trusted.

The role of the media in opinion formation and change does not go unchallenged in the literature. There are those whoargue that the public opinion may not be that easily malleable and that it is in fact stable, and coherent over time (Page andShapiro, 1992; Popkin, 1994; Page and Bouton, 2006). Others argue that the stability that had formerly characterized publicopinion corresponded to the existence of a stable international system under the Cold War and that public opinion in thepost-Cold War era has undergone change in light of uncertainty in the new international realm (Entman, 2004). A cursorylook at the public opinion trends in the US since the beginning of 1990s demonstrate in fact that the Cold War mentality ofseeing Russia as unfriendly has not been a constant. As Fig. 1 shows, Americans’ opinion of Russia has become favorableduring normal times only to become unfavorable after major crises between the two countries. We attribute this pattern ofpublic opinion around and between crises to the role media plays in ratcheting up the Cold War rhetoric. Having an infor-mational disadvantage, people rely on shortcuts provided by the media to interpret the events happening on the ground.

Whether or not this negative media framing and the consequent negative public opinion has had any affect on relationsbetween the two countries is another area where the literature falls short of providing much guidance. There are those whoargue that the media serves as an independent, strategic actor in the policymaking process with an important mission toinform the public (Patterson, 2000). It is argued that especially in the absence of clear national interests and foreign policydoctrines in the post cold-war period, the media has had a much critical role in shaping the foreign policy agenda (Entman,2004). Yet, others claim that the media acts as a conveyer belt and that its reliance on official sources for information oninternational events leads to recycling of the prepackaged frames obtained from decision-makers (Cohen, 1963; Brody, 1991;Malek, 1996; Lehman-Wilzig and Bloch 2002). Similarly, there is no consensus on how public opinion shapes foreign policy.Many would argue that the effect of public opinion depends on the type of foreign policy decisions to be made, the politicalprocess, media goals at the time and perceived relevance of the public opinion among others (Bloch-Elkon, 2007).

We argue that the events of August 2008 have had some direct impact on the bilateral relationship between US and Russia.Moscow’s recognition of the independence of South Ossetia and Abkhazia, its continued military presence in the formerSoviet territories as well as the unrelenting American economic andmilitary support to Georgia’s independence have directlycomplicated the relations between these two countries. But perhaps the indirect impact of the war has been much moreprofound. The fact that this war took place in the midst of a presidential election campaign in the United States and wasportrayed as a foreign policy litmus test for the candidates made Russia a part of the national political conversation. Althoughbrief, the war received considerable media attention in the United States.13 “During the three presidential debates that fall,

13 According to Pew Research Center’s Project for Excellence in Journalism, during the week of August 11 and 17, the war between Russia and Georgia with26% of overall coverage was the biggest story in the US media, marking the first week in nine months when the presidential campaign did not generate themost coverage. ‘War in Georgia is bigger news than the campaign’, PEJ Campaign Coverage Index, 11–17 August 2008, available at http://www.journalism.org

O. Bayulgen, E. Arbatli / Communist and Post-Communist Studies 46 (2013) 513–527 525

Russia was probably the foreign policy topic mentioned most often after the wars in Iraq and Afghanistan” (Charap, 2010,285). Senators John McCain and Barack Obama competed with each other to rebuke Russia’s actions and offer varioussanctions to punish Russia. Barack Obama, who came close to losing his lead during the crisis, responded initially in a moremeasured way, for which he was instantly criticized by many that he is unfit to be Commander-in-Chief. In a poll conductedafter the war in Georgia, 55 percent of likely voters named McCain as best qualified to deal with Russia, compared to 27percent for Obama (Charap, 2010, 286). However, “as the crisis escalated, Obama’s criticism of Russia sharpened. In hisstatement on 11 August, Obama put the blame for the escalation of hostilities squarely on Russia and raised the possibility ofblocking Moscow’s application to join the World Trade Organization (WTO) as well as hinting at the prospect of a possibleboycott of the 2014Winter Olympics to be held in Sochi, near the Georgian border. Like McCain, Obama has also made it clearthat he considers the crisis a ’turning point’ in Russia’s relations with the West” (Zaborowski, 2008).

Such politicization of the foreign policy toward Russia suggests that negative media framing and public opinion can beused by politicians to justify a hawkish and defiant foreign policy rhetoric. Similar trends canmost likely be observed in Russiawhere theories about Western plots to isolate Russia in the world get reinforced in the media, crystallized in public opinionagainst the US and used by elites to ratchet up the ColdWarmentality.14While it is difficult to sort out the exact feedback loopbetween the media, public opinion and foreign policy elites, it is not hard to see the central place the media occupies inconstructing a prism through which the public and elites perceive complex information.

It is important, finally, to acknowledge that the resurgence of the Cold War mentality in the US–Russia relations appearsmore symbolic and ephemeral than instrumental and constant. Since the resetting of relations by the new US administrationin 2009 – not long after the August 2008wardthe dialog and cooperation between the two countries have been improving onissues such as nonproliferation, Afghanistan, bilateral trade and others. This proves that the ColdWar mentality does not leadto a complete breakdown of relations; there are in fact too many other factors that shape the contours of the interactionbetween the two countries. It does, however, have the effect of slowing down negotiations and stalling progress on importantissues. It also provides a short-term political tool for elites to position themselves via their competitors in the domestic realm.We argue that in this recycling ritual of the Cold War mentality, the media plays a complex, yet significant role.

Acknowledgments

The authors would like to thank Timofei V. Bordachev, Maxim V. Bratersky, Jeffrey Ladewig, Sergei A. Medvedev, Andrei Y.Melville, Mikhail G. Mironyuk, Natalia Piskunova and Dmitry V. Suslov for valuable comments and input in this paper.

References

Auerbach, Y., Bloch-Elkon, Y., 2005. Media Framing and Foreign Policy: the Elite Press vis-à-vis US Policy in Bosnia, 1992–95. J. Peace Res. 42 (1), 83–99.Beaudoin, C., Thorson, E., 2002. Spiral of violence? Conflict and conflict resolution in international news. In: Gilboa, E. (Ed.), Media and Conflict. Trans-

national Publishers, New York.Bennett, W.L., Lawrence, R., Livingston, S., 2006. None dare call it torture: indexing and the limits of press independence in the Abu Ghraib Scandal. J.

Commun. 56 (3), 467–485.Berry, O.N., 1990. Foreign Policy and the Press: an analysis of The New York Times Coverage of U.S. Foreign Policy. Greenwood Press, Westport CT, New York.Bissell, K.L., 2002. The crisis in Kosovo: photographic news of the conflict and public opinion. In: Gilboa, E. (Ed.), Media and Conflict. Transnational Pub-

lishers, New York, pp. 311–330.Bloch-Elkon, Y., 2007. Studying the media, public opinion and foreign policy in international crises: the United States and the Bosnian crisis, 1992-1995. The

Harv. Int. J. Press/Polit. 12 (20), 20–51.Brody, R.A., 1991. Assessing the President: The Media, Elite Opinion and Public Support. Stanford University Press, Stanford.CBS News, 2012. Anti-U.S. Rhetoric Heats up Putin Campaign [online] Available from: http://www.cbsnews.com/8301-202_162-57380262/anti-u.s-rhetoric-

heats-up-putin-campaign/ (accessed 28.05.13.).Chang, T., Shoemaker, P.J., Brendlinger, N., 1987. Determinants of international news coverage in the US Media. Commun. Res. 14, 396–414.Chang, T., 1993. The Press and China Policy: the Illusion of Sino-Asmerican Relations, 1950–1984. Ablex Publishing Corporation, Norwood, NJ.Chang, T., 1998. All countries not created equal to be news: world system and international communication. Commun. Res. 25 (2), 528–566.Charap, S., 2010. ‘The transformation of US–Russia relations’. Curr. Hist. (October), 281–287.Chivers, C.J., 2008. In Battered Villages, Georgian Speak, if They Dare. The New York Times [online] 19th August. Available from: http://www.nytimes.com/

2008/08/19/world/europe/19villages.html?pagewanted¼all (accessed 15.09.08.).Chivers, C.J., Barry, E., 2008. Georgia Claims on Russia War Called Into Question. The New York Times [online] 7th November. Available from: http://www.

nytimes.com/2008/11/07/world/europe/07georgia.html?pagewanted¼all (accessed 02.11.08.).Chivers, C.J., 2010. Embracing Georgia, U.S. Misread Signs of Rifts. The New York Times [online] 1st December. Available from: http://www.nytimes.com/

2010/12/02/world/europe/02wikileaks-georgia.html (accessed 17.02.11.).Chong, D., Druckman, J.N., 2007. Framing theory. Annu. Rev. Polit. Sci. 10, 103–126.Cohen, B.C., 1963. The Press and Foreign Policy. Princeton University Press, New Jersey.Denhan, B.E., 1997. Anonymous attribution during two periods of military conflict: using logistic regression to study veiled sources in American newspapers.

Journal. Mass Commun. Quart. 74 (3), 565–578.Dreier, P., 1982. The position of the press in the US power structure. Soc. Probl. 29, 298–310.Ertel, Manfred, Klussmann, U., Koelkl, S., Mayr, W., Schepp, M., Stark, H., Szandar, A., August 31 2008. Road to War in Georgia. Der Spiegel.Entman, R., 1991. Framing US coverage if international news: contrasts in narratives of the KAL and Iran Air incidents. J. Commun. 41, 6–27.Entman, R., 1993. Framing: towards clarification of a fractured paradigm. J. Commun. 43, 51–58.Entman, R., Page, B., 1994. The news before the storm. In: Bennett, W.L., Paletz, David (Eds.), Taken by Storm: the Media, Public Opinion, and U.S. Foreign

Policy in the Gulf War. University of Chicago Press, Chicago.

14 For instance, a Russian poll by the Levada Center in Moscow showed that 74% of Russians polled believed that Georgia was a pawn of the United Statesand asked about the causes of the crisis, 49% cited Washington’s policies in the region, while 32% blamed Georgia Levy (2008b).

O. Bayulgen, E. Arbatli / Communist and Post-Communist Studies 46 (2013) 513–527526

Entman, R.M., 2004. Projections of Power: Framing News, Public Opinion, and U.S. Foreign Policy. University of Chicago Press, Chicago.Entman, R., 2007. Framing bias: media in the distribution of power. Journal of Communication 57, 163–173.FAIR, 2008. Georgia/Russia Conflict Forced into Cold War Frame. Media Advisory, Fairness and Accuracy in Reporting [online] 14th August. Available from:

http://fair.org/take-action/media-advisories/georgiarussia-conflict-forced-into-cold-war-frame/ (accessed 15.09.09.).Gans, H., 1979. Deciding What’s News. Pantheon, New York.Gitlin, T., 1980. The Whole World Is Watching: Mass Media in the Making and Unmaking of the New Left. University of California Press, Berkeley, CA.Golan, G., Wanta, W., 2001. Second-level agenda setting in the New Hampshire primary: a comparison of coverage in three newspapers and public per-

ceptions of candidates. Journal. Mass Commun. Quart. 78, 247–259.Golan, G., Wanta, W., 2003. International elections on US network news. Gaz. Int. J. Commun. Stud. 65 (1), 25–39.Golan, G., 2006. Inter-media agenda setting and global news coverage: assessing the influence of the New York Times on three network television evening

news programs. Journal. Stud. 7 (2), 323–333.Goffman, E., 1974. Frame Analysis: an Essay on the Organization of Experience. Harvard University Press, Cambridge, MA.Gorbachev, M., 2008. Russia Never Wanted a War. The New York Times [online] 20th August. Available from: http://www.nytimes.com/2008/08/20/

opinion/20gorbachev.html (accessed 15.05.09.).Groeling, T., Baum, M., 2008. Crossing the water’s edge: elite rhetoric, media coverage, and the rally-round-the-flag phenomenon. J. Polit. 70, 1065–1085.Hallin, D.C., 1994. We Keep America on Top of the World. Routledge, New York.Hahn, G.M., 2008. The Making of Georgian–Russian Five-day August War: A Chronology. June–August 8, 2008. Russia: Other Points of View. [online]

Available from http://www.russiaotherpointsofview.com/gordon-hahns-underground-.html (accessed: 10.03.10.).Hatchen, W.A., Scotton, J.F., 2007. The World News Prism: Global Information in a Satellite Age. Blackwell Publishing.Herman, E., 1993. The media’s role in US foreign policy. J. Int. Aff. 47, 23–45.Hess, S., 1996. International News and Foreign Correspondents. The Brookings Institution, Washington, D.C.Holsti, O.R., 1969. Content Analysis for the Social Sciences and Humanities. Addison-Wesley Publishing Company, Massachusetts.Keeter, S., Miller, C., Kohut, A., Groves, R.M., Presser, S., 2000. Consequences of reducing nonresponse in a national telephone survey. Pub. Opin. Quart. 64,

125–148.King, Charles, 2008. The five-day war: managing Moscow after the Georgian crisis. Foreign Aff. 87 (6), 2–11.Krippendorff, K., 2004. Content Analysis: an Introduction to Its Methodology. Sage Publications, Thousand Oaks.Larson, J.F., 1984. Television’s Window on the World: International Affairs Coverage on the US Networks. Ablex Publishing, Norwood, NJ.Lee, J., Craig, R.L., 1992. News as an ideological framework: comparing US newspapers’ coverage of labor strikes in South Korea and Poland. Discourse Soc. 3

(3), 341–363.Lehman-Wilzig, S., Bloch, Y., 2002. An Exploratory Model of Media-Government Relations in International Crisis: U.S. Involvement in Bosnia. In: Gilboa, E.

(Ed.), Media and Conflict: Framing Issues, Policy Making, Shaping Opinion. Transnational Publishers, New York, pp. 1922–1995.Levy, C.J., 2008a. Russia Prevailed on the Ground, but Not in the Media. The New York Times [online] 22th August. Available from: http://www.nytimes.com/

2008/08/22/world/europe/22moscow.html (accessed 15.09.08.).Levy, C.J., 2008b. Russia Adopts Blustery Tone Set by Envoy. The New York Times [online] 28th August. Available from: http://www.nytimes.com/2008/08/

28/world/europe/28moscow.html?pagewanted¼all&_r¼0 (accessed 15.09.08.).Lévy, B., 2008. Russia is Brazen, EuropeWeak. Wall Street J. [online] 27th August. Available from: http://online.wsj.com/article/SB121979710251174853.html

(accessed 15.09.08.).Lobjakas, A., 2009. EU Report on 2008War Tilts against Georgia. Radio Free Europe Radio Liberty [online] 30th September. Available from: http://www.rferl.

org/content/EU_Report_On_2008_War_Tilts_Against_Georgia/1840447.html (accessed 10.03.10.).Lombard, M., Snyder-Duch, J., Bracken, C.C., 2002. Content analysis in mass communication: assessment and reporting of intercoder reliability. Human

Communication Research 28, 587–604.Malek, A., 1996. ‘New York Times’ editorial position and U.S. foreign policy: the case of Iran revisited’. In: Malek, A. (Ed.), News Media and Relations: a

Multifaceted Perspective. Ablex Publishing Corporation, NJ.Malek, A., Wiegand, K.E., 1996. News media and foreign policy: an integrated review. In: Malek, A. (Ed.), News Media and Foreign Relations: a Multifaceted

Perspective. Ablex Publishing Corporation, NJ.McCombs, M.E., Shaw, D.L., 1972. The agenda-setting function of mass media. Pub. Opin. Quart. 36 (2), 176–187.McCombs, M.E., Llamas, J.P., Lopez-Escobar, E., Rey, F., 1997. Candidate images in Spanish elections: second level agenda-setting effects. Journal. Mass

Commun. Quart. 74, 703–717.McLeod, J., Pan, Z., Rucinski, D., 1995. ‘Level of Analysis in Public Opinion Research’, Public Opinion and the Communication of Consent. Guilford Press, New

York.McNelly, J.T., Izcaray, F., 1986. International news exposure and images of nations. Journal. Quart. 63 (3), 546–553.Merrill, J.C., Fisher, H.A., 1980. The World’s Greatest Dailies. Hastings House, New York.Merrill, J.C., 1995. Global Journalism. Longman, New York.Nelson, T.E., Clawson, R.A., Oxley, Z.M., 1997. Media framing of a civil liberties conflict and its effects on tolerance. Am. Polit. Sci. Rev. 91, 567–583.Neuendorf, K., 2002. The Content Analysis Guidebook. Sage Publications, Thousand Oaks, CA.Oppel, R.A., 2012. Romney’s Adversarial View of Russia Stirs Debate. The New York Times [online] 11th May. Available from: http://www.nytimes.com/2012/

05/12/us/politics/romneys-view-of-russia-sparks-debate.html?pagewanted¼all&_r¼0 (accessed 28.05.13.).Page, B.I., Shapiro, R.Y., 1992. The Rational Public. University of Chicago Press, Chicago.Page, B.I., Bouton, M.M., 2006. The Foreign Policy Disconnect: What Americans want from our Leaders but don’t get. University of Chicago Press, Chicago.Paletz, D.L., 2002. The Media in American Politics. Longman, New York.Park, J., 2003. Contrasts in the coverage of Korea and Japan by US television networks: a frame analysis. Gazette 65, 145–164.Price, V., Tewksbury, D., 1995. News values and public opinion: a process: how political advertising and TV news prime viewers to think about issues and

candidates. In: Biocca, F. (Ed.), Television and Political Advertising. Lawrence Erlbaum Associates, Hillside, NJ.Patterson, T.E., 2000. How Soft News and Critical Journalism are Shrinking the News Audience and Weakening Democracy. Joan Shorenstein Center on the

Press, Politics, and Public Policy. Kennedy School of Government, Harvard University, Cambridge, MA.Popkin, S., 1994. The Reasoning Voter, second ed. University of Chicago Press, New York.Protess, D., McCombs, M., 1991. Agenda Setting: Readings on Media, Public Opinion, and Policymaking. Lawrence Earlbaum, Hillsdale, NJ.Reilly, J., 1979. American Public Opinion and U.S. Foreign Policy. Chicago Council on Foreign Relations, Chicago.Reese, S.D., Danielian, L.H., 1989. Intermedia influence and the drug influence: converging on Cocaine. In: Shoemaker, P. (Ed.), Communication Campaigns

about Drugs: Government, Media, Public. Lawrence Earlbaum, Hillsdale, NJ, pp. 29–46.Reese, S.D., 2001. ‘Prologue-framing life: a bridging model for media research’. In: Reese, S.D., O.Gandy, Grant, A.E. (Eds.), Framing Public Life: Perspectives

on Media and Our Understanding of the Social World. Lawrence Erlbaum, Mahwah, NJ, pp. 7–31.Riffe, D., Lacy, S., Fico, F.G., 2005. Analyzing Media Messages: Using Quantitative Content Analysis in Research. Lawrence Erlbaum Associates, New Jersey.Robinson, P., 2000. The policy-media interaction model: measuring media power during humanitarian crisis. J. Peace Res. 37 (5), 613–633.Salwen, M.B., Matera, F.R., 1992. Public salience of foreign nations. Journal. Quart. 69 (3), 623–632.Semetko, H.A., Brzinski, J.B., Weaver, D., Willnat, L., 1992. TV news and US public opinion about foreign countries: the impact of exposure and attention. Int.

J. Pub. Opin. Res. 4 (1), 18–36.Sestanovich, Stephen, 2008. What has Moscow done? Rebuilding US–Russian relations. Foreign Aff. 87 (6), 12–28.Shoemaker, P., Reese, S., 1996. Mediating the Message: Theories of Influences on Mass Media Content, second ed. Longman, White Plains, NY.

O. Bayulgen, E. Arbatli / Communist and Post-Communist Studies 46 (2013) 513–527 527

Sigal, L.V., 1973. Reporters and Officials: the Organization and Politics of Newsmaking. D.C. Health and Co, Lexington, Massachusetts.Sniderman, P., Brody, R., Tetlock, P., 1991. Reasoning and Choice: Explorations in Political Psychology. Cambridge University Press, New York.Soroka, S.N., 2003. Media, public opinion, and foreign policy. Press/Polit. 8 (1), 27–48.Stevenson, R.L., Shaw, D.L., 1984. Foreign News and the New World Information Order, first ed. Iowa State University Press, Ames.Tedesco, J.C., 2001. Issue and strategy agenda-setting in the 2000 presidential primaries. Am. Behav. Sci. 44 (12), 2048–2067.Tuchman, G., 1978. Making News: a Study in the Construction of Reality. Free Press, New York.Tsygankov, A.P., Tarver-Wahlquist, M., 2009. Duelling honors: power, identity and the Russia–Georgia divide. Foreign Policy Analysis 5 (4), 307–326.Tsygankov, A.P., 2010. Russia’s Foreign Policy: Change and Continuity in National Identity. Rowman & Littlefield Publishers, Lanham.Viser, M., 2003. Attempted objectivity: an analysis of the New York Times and Ha’aretz and their portrayals of the Palestinian–Israeli conflict. Press/Polit. 8,

114–120.Wanta, W., Hu, Y., 1993. The agenda-setting effects of international news coverage: an examination of differing news frames. Int. J. Pub. Opin. Res. 5, 250–

264.Wanta, W., Golan, G., Lee, C., 2004. Agenda setting and international news: media influence on public perceptions of foreign nations. Journal. Mass

Commun. Quart. 81, 364–377.Weiss, C.H., 1974. What America’s leaders read. Pub. Opin. Quart. 38, 1–22.Wilhoit, G.C., Weaver, D., 1983. Foreign news coverage in two US wire services: an update. J. Commun. 33 (2), 132–148.Wu, H.D., 1998. Investigating the determinants of international news flow: a meta-analysis. Int. J. Commun. Stud. 60, 493–512.Zaborowski, M., 2008. Back to the Future? US–Russia Relations after Georgia [online]. ISS European Union Institute for Security Studies, 2nd September.

Available from: http://www.iss.europa.eu/publications/detail/article/back-to-the-future-us-russia-relations-after-georgia/ (accessed 23.05.13.).Zaller, J., Chiu, D., 2000. Government’s little helper: U.S. press coverage of foreign policy crises, 1946–1999. In: Nacos, B.L., Shapiro, R.Y., Isernia, P. (Eds.),

Decisionmaking in a Glasshouse. Rowman & Littlefield, New York.Zelizer, B., Park, D., Gudelunas, D., 2002. ‘How bias shapes the news: challenging the New York Times’ status as a newspaper of record on the Middle East’.

Journalism 3, 283–307.