University of Amsterdam Graduate School of Communication
Erasmus Mundus Master's in Journalism, Media and Globalisation
Differences in the Ukrainian crisis coverage in Lithuanian and Russian TV news broadcast in Lithuania
Master’s Thesis
Gabrielė Gedvilaitė Student number: 11081805 Date of submission: 27-05-2016 Supervisor: Dr. Andreas R. T. Schuck
Abstract
This study examines the differences in the Ukrainian crisis coverage in Lithuanian TV news and Russian TV news broadcast in Lithuania. The focus of the study is the extent to which visual and textual frames varied in three media outlets. The analysis also includes the examination of sources and their evaluation in the context of the Ukrainian crisis. A content analysis of the evening newscasts in one Russian TV channel (which is broadcast in
Lithuania) and two Lithuanian TV channels is conducted. Findings reveal that the beginning of the crisis was covered similarly in Lithuanian and Russian news, though it changed over time. This study also shows that the biggest change in framing the crisis over time was in the Russian news, while there were fewer changes in Lithuanian news. The Russian news focused on individuals and emotional coverage. Such coverage was also present in Lithuanian news, together with the conflict frame. The analysis of sources shows that the same types of actors dominated the coverage of the conflict in all three media outlets, though their position towards the conflict and their evaluation was different among chosen outlets. The analysed media outlets are broadcast in Lithuania therefore, Lithuanian audiences were exposed to a different coverage of the Ukrainian crisis, and further research is needed to analyse how such
differences might affect the audience.
The current study contributes to the scholarly gap of the representation of the Ukrainian crisis in the media and also provides empirical evidence that can be used in discussions of the Russian media export to other countries.
Introduction
In April 2015, the Radio and Television Commission of Lithuania terminated the broadcasting of Russian TV channel RTR Planeta in Lithuania. The decision was based on violations of the Public Information Act, specifically for the incitement of war and hate in the context of the Ukrainian crisis (The Radio and Television Commission of Lithuania, 2015). Between 2013 and 2015, there were 6 cases when Lithuanian media watchdogs suspended Russian TV channels for similar violations (The Radio and Television Commission of Lithuania, 2015).
Russian journalistic productions disseminated in Lithuania have to obey local
legislation of information in the media; however, the cases of violations show that the content is created based on Russian journalistic norms, which include low journalistic
professionalism, siding with the Kremlin’s position and using media as an information war tool (Koltsova, 2001; Becker, 2004; Lipman, 2005). Even though there has not been any analysis on what the effects are of being exposed to Russian media in other countries, concerns over such content can be based on research which shows that the Russian
government has used media export as an information war tool in post-Soviet societies before engaging in military conflicts, such as wars in Ukraine and Georgia (Martišius, 2010;
Martišius, 2014; Szostek, 2014).
Russian TV in Lithuania primarily targets a Russian ethnic minority which makes up 6% of the general population (Statistics Lithuania, 2011), though a third of Lithuanians claim to often watch Russian television (Nevinskaitė, 2009). In addition, Russian media is the most popular foreign media in Lithuania and television is the leading type of media (Nevinskaitė, 2009). Such data suggest that a significant number of Lithuanians are exposed to Russian journalistic production which has a potential to influence and shape public opinion.
Thus far, the topic of Russian media content has been researched to a limited extent. Several studies have focused on Russian propaganda and information strategy in the context of the Ukrainian war (Mažeikis, 2014; Dobre, 2015; Hansen, 2015; Suslov, 2014; Teper, 2015). A certain amount of scholarly attention has been given to the analysis of media
coverage of previous conflicts where Russia was involved, such as the war in Georgia in 2008 (Heinrich & Tanaev, 2009; Eckler & Kalyango, 2012; Baysha, 2010) or the Chechen wars (Malinkina & Douglas, 2000). However, literature review suggests that media coverage of one of the most recent conflicts involving Russia – the Ukrainian crisis – has been analysed to a very limited extent.
A review of the relevant literature shows that there has not been any analysis on the presentation of the Ukrainian crisis in Lithuanian media. In addition, literature on the
differences between Lithuanian and Russian media content disseminated in Lithuania, and the effects of these differences, is rather normative and lacks empirical evidence. Available studies provide only theoretical discussion on such effects and no research-based studies could be found.
The current study presents a comparative analysis of the extent to which coverage of the Ukrainian crisis differed between Lithuanian TV channels and Russian TV channels broadcast in Lithuania. By using framing theory as a baseline, the content analysis reveals differences in terms of frames and sources used in the coverage. This study also sheds light on the importance of visual analysis, which has been lagging behind textual analysis in previous research. The results of this study provide empirical data that can be used by media
organisations, politicians or society when discussing the issues related to Russian media export to other countries. In addition, this study contributes to research on the Ukrainian crisis by providing more knowledge on the coverage in two countries.
Theoretical framework Framing in the media
The literature review on the concept of framing shows that there are several aspects that should be emphasized when defining framing. Firstly, frames are related to the process of selection and salience: “to frame is to select some aspects of a perceived reality and make them more salient in a communicating text, in such a way as to promote a particular problem definition, causal interpretation, moral evaluation, and / or treatment recommendation for the item described” (Entman, 1993, p. 52). Therefore, framing is a choice in what and how to communicate. The aspect of salience can be reinforced by repeating the same objects, words or symbols which eventually can be interpreted in the same way (Entman, Matthes &
Pellicano, 2008; Reese, Gandy & Grant, 2001).
Secondly, framing is related to the aspect of time as the same message can become symbolic if it is communicated routinely over a concentrated period of time (Entman, Matthes & Pellicano, 2008; Reese, Gandy & Grant, 2001). This means that the presence of frames in the media does not guarantee its influence on the audience if it is not transmitted over a certain period of time.
Finally, the effects of frames are also related to the social and cultural context they act in (Entman, 1993; Reese, Gandy & Grant, 2001). For frames to have impact, they need to be shared among the members of the society and to resonate with their cultural and social practices (Nisbet, 2010). This implies that there should be a connection between the frames and an individual’s prior knowledge, and also that frames might have different impacts across different cultures and societies.
In the field of mass communication, one of the main transmitters of frames is
considered to be the media. They have a double role: media can be seen as a passive provider or just an arena of frames, but they can also be understood as a platform where certain frames
are constructed (Delli-Carpini, 2005). This construction can take place on the macro level, where journalists make a choice of the means to present the information (Maier & Rittberger, 2008). Such frame-building (de Vreese, 2005) can be related to the routines and practices of media organization. Media scholars assume that it is impossible for journalists to avoid
framing as reporters have to tell a story within a limited amount of time or space (Valkenburg, Semetko & de Vreese, 1999).
At the micro level, the media transmits the frames to the audience, which perceives and interprets the information in a certain way (Maier & Rittberger, 2008). This process can be defined as frame-setting (de Vreese, 2005) in which the media interacts with the audience through framing and connects certain messages, issues, or concepts to individuals’ prior knowledge, as well as cultural context (Nisbet, 2010).
Recently, a number of researchers have mainly analysed two types of frames: generic or issue-specific. Generic news frames are defined as relatively universal and not related to any specific topic or issue; their main quality is that they can be used for comparison (de Vreese, 2003). Issue-specific frames can be applied only to specific problems or events and even though they cannot be used to distinguish general patterns, they allow to capture issue-specific details (de Vreese, 2003).
In recent decades, the theory of framing has been extensively used in order to analyse media content, and since Entman’s attempt to provide a more universal concept of frames (1993), the number of framing studies has increased significantly (Vliegenthart & van
Zoonen, 2011). Framing studies have also received some criticism over the lack of a coherent definition (Cacciatore, Scheufele & Iyengar, 2016; Corbu & Iorgoveanu, 2012; Vliegenthart & van Zoone, 2011) and the inconsistent methodological approaches that have been used to analyse frames (Entman, Matthes & Pellicano, 2008; Matthes & Kohring, 2008; Golčevski, Von Engelhardt & Boomgaarden, 2013).
Another point of criticism is related to the fact that research of frames in the media has mainly focused on the analysis of text while visual examination has received only sporadic attention (Anden-Papadopolous, 2008; Edy & Meirick, 2007; Coleman, 2010; Coleman & Banning, 2006; Greenwood & Jenkins, 2015; Messaris & Abraham, 2001; Grabe & Bucy, 2009; Bucy & Grabe, 2007; Esser, 2008).
Scholars claim that visuals might tell an additional or a different story than the verbal message (Perlmutter & Wagner, 2004; Graber, 1996; Coleman, 2010). Therefore, exclusively textual analysis might not show the overall picture of the frames. However, some researchers criticize such approach and point out that the mismatch between the visual and verbal story does not always occur in the news and that visuals are often closely related to the textual information (Mellese & Müller, 2012; Powell, Boomgaarden, De Swert & de Vreese, 2015; Gamson, 1989; Griffin, 2004). Review of such studies shows that a clear conclusion of the extent to which visuals reflect verbal messages cannot be made, as these studies vary greatly in the chosen topics, methodology, analysed media and form of visuals.
Even though the existing research of visual frames is less prominent than the research of textual frames, analysis of both types of news presentation could contribute to a better understanding of frames in the media. As noted in the literature, frames consist of a combination of visual and textual content, therefore visuals should be considered as an important part of the overall message that the media are communicating (Corbu &
Iorgoveanu, 2012; Graber, 1996; Powell, Boomgaarden, De Swert & de Vreese, 2015). To include the analysis of visuals is also important from the audience point of view, as they process visual and verbal messages simultaneously (Coleman, 2010; Coleman & Banning, 2006; Bucy & Grabe, 2007). For these reasons, both forms of news presentation should be analysed in order to get a complete insight of frames in the news.
Implications of framing
Literature suggests that the audience derives the meaning on certain topics from the media and it is crucial to analyse how they structure the information (Boaz, 2005; Tankard, 2001; Tannen, 1993; Iyengar, 1991), because the media has power to shape public
understanding, knowledge and interpretation of the events (McNair, 2003). The media does not simply present the facts, it also gives important cues on how the news event should be understood (Schnell & Callaghan, 2005).
Media effects theory analyses frames as one of the tools that media use to simplify complex issues by emphasizing certain elements over the others (Nisbet, 2010; Iyengar, 1991). A number of studies in the field of political communication showed how various frames affect attitudes and opinions on different topics, issues or problems (e.g. Lecheler & de Vreese, 2012; Boukes et al., 2015; Lecheler, Schuck & de Vreese, 2013; Schuck & de Vreese, 2006).
Furthermore, scholars who incorporate visual analysis in their framing studies emphasize the important framing effects of visual content as well. It is argued that visual images lend greater credibility, as seeing the event leads to believing (Messaris & Abraham, 2001; Graber, 1996; Fahmy, 2010; Corbu & Iorgoveanu, 2012) which implies that the
recipient may be easily persuaded by visual content. In addition, studies show that imagery is particularly powerful when emotional or controversial topics are presented (Powel,
Boomgaarden, De Swert & de Vreese, 2015; Graber, 1996; Coleman, 2010). For example, visuals in the coverage of war, conflict or terror often evoke negative emotions, fear and a feeling of vulnerability, which are believed to have even stronger effects on the audience (Gadarian, 2010; Gadarian, 2014; Brader 2006). Furthermore, some studies have also shown that visual information has longer recall effects than textual messages (Gibson & Zillmann, 2000; Zillmann, Gibson & Sargent, 1999; Powel, Boomgaarden, De Swert & de Vreese,
2015), which also suggests that visual frames might have stronger media effects. Such findings demonstrate that it is essential to include visual analysis in order to understand framing effects better.
Framing war and conflict and the role of sources
The coverage of war and conflict is one of the research areas that framing studies have focused on in the past. Research shows that media coverage during war is often patriotic and uncritical (Aday, Cluverius & Livingston, 2005; Kellner, 1992; Newhagen, 1994) and the narrative of war is usually in line with the official or public discourse and narratives (Griffin, 2004; Dimitrova & Strömbäck, 2005; Sporer-Wagner, 2005). The implication of such
coverage lies in the understanding of how uncritical or biased view affects the public as the coverage of war not only informs, but also shapes the way the audience makes sense of the conflict (Schwalbe, 2006).
Comparative studies of war and conflict presentation also show that the same war is not covered in the same way across different countries and media systems (Fahmy, 2010; Dimitrova & Strömbäck, 2005). Such differences might result in critical perceptions of the other and reinforce the polarization in worldwide public opinion (Fahmy, 2010; Dimitrova & Strömbäck, 2005).
One of the important elements that influences war coverage and its framing effects is the choice of sources (Dimitrova & Connolly-Ahern, 2007; Strömbäck et al., 2013; Lawrence, 2000; Berkowitz, 1987; Manning, 2001). News sources influence news content, as the
selection of who gets to speak has an impact on what message is presented in the news (Strömbäck et al., 2013; Kim & Weaver, 2003). Therefore, it is not only a journalist, but also a source who contributes to the frame-building of news events.
The analysis of news sources is important not only because it allows to understand the existing frames in the news, but also because it influences audiences’ understanding of certain
issues. No news organization can communicate all sides of the conflict because of the obstacles that war reporters face; however, a wide variety is important because it shapes public’s knowledge and opinion on the issue (Carpenter, 2007; Cook, 1998; Manning, 2001). Therefore, it is important to include the analysis of sources, as frames can be seen as a coproduction of journalists and their sources.
One of the main findings of previous research on the use of sources in war coverage is that journalists tend to rely on official sources more often than any other sources, and this is recurring across countries (Dimitrova & Connolly-Ahern, 2007; Strömbäck et al., 2013; Lawrence, 2000; Manning, 2001; Dimitrova & Strömbäck, 2009a; McQuail, 2010). The coverage of war and conflict is also often dominated by military sources (Dimitrova & Strömbäck, 2009a; Carpenter, 2007; McQuail, 2010). One of the reasons for this is the limits of accessibility to a more diverse range of sources, especially when the conflict takes place in another country (Carpenter, 2007). Additionally, poor understanding of military culture and war might influence the selection of sources (Carpenter, 2007). Finally, the political context in which the conflict is covered also has impact as war reporters might experience external pressure if their own country is involved in the military conflict (Dimitrova & Strömbäck, 2009a; Seib, 2002).
Features of Lithuanian and Russian media
For the reasons mentioned in the previous section, it is important to take into
consideration the media systems chosen for this analysis. Even though this study focuses on the media content in Lithuania, the analysed Russian news are produced in Russia and then broadcast in Lithuania (Martišius, 2014). Therefore, it is important to discuss media systems in both countries.
Studies on the media landscape in Russia stress a high level of state control over the media content, with a strictly limited diversity of opinions related to the issues of regime, elections and international relations (Layton, 2014; Becker, 2004; Lipman, 2005). The lack of
pluralism and competitiveness in the media market, absence of opposition to the Kremlin’s official viewpoints, and a deficit of alternative media imply that consumers of Russian media are exposed to one-sided information which echoes the official voices of the government (Eniklopov, Petrova & Zhuravskaya, 2011).
The Lithuanian media landscape differs significantly from the Russian one. Several analyses show that Lithuanian media system shares a lot of features with media systems in Western democracies, such as diverse media ownership, strong commercialization or sufficiently high media pluralism (Balčytienė, 2012; Jastramskis, 2011). On the contrary, literature suggests that features of Russian media system has a lot in common with authoritarian media systems (Becker, 2004; Toepfl, 2011). Therefore, when comparing Lithuanian and Russian media systems it can be stated that they are essentially different, as journalistic content is produced under different circumstances and is influenced by different factors.
The Ukrainian crisis in the media
The events in Ukraine started taking place shortly after the Eastern Partnership Summit in 2013 (which was organised during the EU Council Presidency held by Lithuania), and it triggered a significant amount of scholarship analysing geopolitical circumstances of the Ukrainian crisis (Jurkynas & Daukšaitė, 2014; Šešelgytė, 2013; Vilpišauskas, 2014; Vilpišauskas, Vandecasteele & Vaznonytė, 2013). However, no analysis on the Lithuanian media coverage of the Ukrainian crisis could be found.
There has been a limited amount of research done on the representation of the Ukrainian crisis in the Russian media. Studies show that the coverage was extremely
prominent in Russian media and it received way more attention than domestic issues (Teper, 2015). Moreover, the issue was highly dichotomized, showing the struggle between right and wrong, or good versus bad (Hansen, 2015). Additionally, Russian media reinforced solidarity
and the need to protect neighbouring country (Hansen, 2015; Teper, 2015) and certain vocabulary and phrases were repeatedly used when representing the conflict (Teper, 2015). Finally, the position of Russian media towards the issue reflected the official stance of Russian government and did not present any alternative viewpoints (Teper, 2015).
Similar findings are presented in the analysis of previous military conflicts that
directly involved Russia. The coverage of the Russia-Georgia conflict in 2008 was also in line with the official position and actions of the Russian government (Baysha, 2010; Heirich & Tanaev, 2009; Eckler & Kalyango, 2012) and the news media turned to official government sources to report on the issue (Basilaia, McKeever & Shaw, 2013). Therefore, the coverage was highly partisan and did not include any alternative views.
The existence of Russian TV in Lithuania implies that Lithuanian audiences are also exposed to one-sided coverage. Literature suggests that in order to gain more support, Russia has been actively strengthening its media outlets abroad, focusing primarily on ex-Soviet countries (Martišius, 2014; Szotek, 2014). Such media outlets were used as soft power tools for disinformation and propaganda purposes in Georgia and Ukraine before Russia engaged in military conflicts there (Martišius, 2010; Martišius, 2014; Szostek, 2014). However, it is difficult to determine the broader implications of Russian media export to post Soviet countries, as existing research lacks data-based findings.
The current study aims to present empirical evidence of the extent to which the coverage of the Ukrainian crisis differed in Russian and Lithuanian news, taking into
consideration the fact that both types of news are accessible to Lithuanian audiences and has the potential to shape public opinion and views on the conflict.
Research questions and hypotheses
Overall research question: To what extent does the coverage of the Ukrainian crisis differ in Lithuanian and Russian TV news broadcast in Lithuania?
Sub question 1: To what extent does the framing of the Ukrainian crisis differ in Lithuanian and Russian TV news broadcast in Lithuania?
Sub question 2: To what extent does the presentation and the use of sources differ in Lithuanian and Russian TV news broadcast in Lithuania?
The following hypotheses are formulated, taking into consideration previous empirical findings on the use of sources in media coverage:
Hypothesis 1: The official sources will be the most dominant group of sources in both Lithuanian and Russian news.
Hypothesis 2: Incumbent governmental sources will dominate the coverage of the Russian news.
Hypothesis 3: The majority of sources presented in the Russian media will defend, justify or agree with Russia’s position, actions or stance on the Ukrainian conflict.
Methods
In order to assess the coverage of the Ukrainian crisis in Lithuanian and Russian TV news, a content analysis is conducted. This method is chosen as it enables the quantification of media content in a systematic and replicable way (Bryman, 2012).
Design of content analysis
The current study analyses the content of the television news as it is the most popular type of news media in Lithuania (Eurobarometer, 2014). In addition, the exposure of the Lithuanian audience to Russian media content is higher for television news than any other type of Russian media (Nevinskaitė, 2009). Research shows that 33% of Lithuanians often watch Russian television, and 44% of Lithuanians indicated that they watch Russian TV sometimes (Nevinskaitė, 2009). Such results also show that Russian television is the most popular media in foreign language in Lithuania compared to other foreign media, such as Polish, English or German.
Three major TV newscast programmes are included in the analysis. For the Lithuanian news, a TV newscast programme TV3 žinios of commercial TV3 channel is selected as it is the most popular newscast in Lithuania (TNS, 2015). Additionally, the newscast Panorama of the Lithuanian public broadcaster LRT is included in order to provide more generalizable data. For Russian TV news, the newscast Vremya of the most popular Russian TV channel in Lithuania, Pervyj Baltijskij Kanal (PBK) (TNS, 2015), is analysed. Even though PBK is a special version of the most watched TV channel in Russia, Perviy Kanal (TNS, 2016), the news programme Vremya consists of the same content and is broadcast from its headquarters in Russia (Martišius, 2014).
Figure 1. The timeline of the events in Ukraine (December 2013 – August 2014)
Period of study. News reports from 1 December 2013 to 17 March 2014 are included in the study as they cover the beginning of the Ukrainian crisis. Major events related to the
Ukrainian crisis took place during this period (see Figure 1): the beginning of protests in Kiev, the escape of Ukrainian president Yanukovych, the take-over of government buildings in Crimea by pro-Russia gunmen, and the Crimean referendum. Additionally, almost one month, from 10 July to 7 August 2014, is included as two important events took place: the
flight MH17 carrying 298 civilians was downed by pro-Russia fighters in the east of Ukraine and the EU sanctions on Russia were imposed.
Data collection. The news reports were collected from the online archives of TV channels and all stories that mentioned ‘Ukraine’ or any other city or region in the country and referred to the ‘events’, ‘crisis’, ‘conflict’, ‘protests’, ‘violence’ or similar were included in the sample (a full list of criteria is presented in Appendix 1). The unit of data collection was the
individual news item, which was defined by the topic. A clear topic change indicated the start and the end of the news item. The news item included not only the report itself, but also its presentation in the studio by the TV news anchor. As a result, 607 news reports (around 40 hours of coverage) fit the criteria and were analysed from all 3 TV channels. The biggest amount of reports was in Russian PBK (n = 284), while there were fewer reports in Lithuanian LRT (n = 196) and TV3 (n = 127).
Measures and operationalisation of content analysis
The coverage was examined by analysing textual and visual framing, and the use of sources. A separate codebook was developed and used during the coding process (see
Appendix 1). The codebook consists of a set of questions that were determined using existing literature. Coding was conducted by the author of this study, who is a native Lithuanian speaker and speaks Russian as a foreign language. In order to ensure the reliability of the study, a randomly selected subset of the full Lithuanian news sample was coded by the second coder and an inter-coder reliability test was conducted using Krippendorff’s alpha coefficient. The results are presented in further sections of this research.
Textual and visual media frames. Framing analysis was carried out using deductive and inductive approaches. The presence and absence of certain generic frames were investigated deductively using operationalisation from previous framing studies. Additionally, one issue-specific frame was developed using an inductive and emerging coding approach based on an
initial investigation of the content of the news coverage by the researcher. In total, ten different frames were applied for the analysis of textual frames.
Generic frames of (1) responsibility, (2) human interest, (3) conflict, (4) morality and (5) economic consequences were operationalised using the measures developed by Semetko and Valkenburg (2000). As the current study analyses the coverage of the Ukrainian war, frames related to conflict were developed using previous studies. The (6) diplomacy and (7) militaristic frames were introduced based on the study of Boaz (2005) and the (8) violence and loss of war and (9) rescue frames were adapted and operationalised using the studies of Hansen (2015), Schwalbe (2006), Gartner (2011), Carpenter (2007), Dimitrova and
Strömbäck (2005).
Using an inductive coding approach of a randomly selected set of news items, a new frame of (10) threat was introduced in this study. Previous research on the coverage of conflicts and terror has shown that war coverage often features threatening content that provokes negative emotions, such as fear or vulnerability (Gadarian, 2010; Gadarian, 2014). In accordance with such findings, the current study developed a threat frame in the context of the Ukrainian war. Three items measuring the threat frame were generated as a result of inductive content analysis in order to indicate whether a news item mentioned that (1) other countries are in danger because of the conflict in Ukraine, (2) there is an intention or
possibility to inflict pain, injury, damage or other hostile action on individuals outside of Ukraine, or that (3) other countries or individuals should be worried about their national security and / or freedom.
Literature review showed that it is unclear whether visuals reflect the textual content or provide additional information to a news story, therefore visual and textual frames were kept similar in this study in order to provide a possibility to compare findings. Just as in textual framing, the analysis of visuals measured the presence and absence of each frame.
As the operationalisation of visual framing in media content is not as standardised as operationalisation of textual framing, several studies of visual war framing were used in order to develop appropriate measures. Generic frames of human interest and conflict were
operationalised based on the study of Semetko and Valkenburg (2000) as it offers universal measures that can be implemented in visual analysis as well. In addition, the militarism frame (based on Boaz, 2005; Gartner, 2011), the diplomacy frame (based on Boaz, 2005;
Greenwood & Jenkins, 2015) and the violence and loss of war frame (based on Gartner, 2011; Dimitrova & Strömbäck, 2005; Carpenter, 2007) were used for the visual analysis. The rest of the textual frames were not applied for the visual analysis as their measures focused mainly on textual messages and developing different measures might have resulted in inconsistency of the use of frames. Therefore, only universal frames were considered for the implementation in the visual analysis.
Frames used in data analysis. A principal component factor solution with Varimax rotation resulted in six distinct factors (see Appendix 2) for textual frames with an eigenvalue of each factor above 1, accounting for 61.19% of the total variance. An examination of the Kaiser-Meyer Olkin measure of sampling adequacy was .695, suggesting that the sample was
factorable. The factor analysis showed that the human interest frame and the violence and loss of war frame empirically constitute one joint frame in the present analysis; because each item of these frames was related to emotions, feelings or effects on individuals, they were
combined into one component named ‘emotions frame’ for the subsequent analysis. In addition, the militarism frame, the threat frame, the economic frame, the conflict frame and the diplomacy frame each constituted separate distinct factors. The frames of responsibility, morality and rescue, which were initially applied in content analysis, were excluded from further discussion as their items did not constitute distinct factors.
The items of the factors included in this research were tested for scale reliability resulting in sufficiently high values: the emotions frame (α = .83), the militarism frame (α = .79), the threat frame (α = .79), the economic frame (α = .75), the conflict frame (α = .71) and the diplomacy frame (α = .55).
The items of the visual frames were also examined using principal component factor solution with Varimax rotation, which resulted in three distinct components (see Appendix 3) with an eigenvalue of each factor above 1, accounting for 63.80% of the total variance. The sample was also factorable as the Kaiser-Meyer Olkin measure of sampling adequacy was .744. Similarly to the textual factors’ solution, items of human interest and violence and loss of war frames clustered into one factor, renamed as emotions frame in this study. However, the item that visually examined the presence of violent acts was initially linked with the violence and loss of war frame, though factor solution showed that it clusters better with the items of the conflict frame. As the analysis of visual frames is more explorative, this
particular item is considered to be a part of the conflict frame further in the analysis.
The visual diplomacy frame was omitted from further analysis as it did not constitute a distinct factor. The items of the three factors were tested for scale reliability and demonstrated sufficiently high values: the emotions frame (α = .79), the conflict frame (α = .76) and the militarism frame (α = .72).
In addition, all frames received sufficient inter-coder reliability scores. The value of the items constituting the textual conflict frame ranged from α = .73 to α = 1.00, the value of the items of the economic frame ranged from α = .69 to α = 1.00, the value of the items of the textual emotions frame ranged from α = .79 to α = 1.00, the value of the items of the visual conflict frame ranged from α = .65 to α = .73 and the value of the items of the diplomacy frame ranged from α = .65 to α = .86. Even though Krippendorff’s alpha could not be calculated for certain frames due to a low variance in the randomly selected test sample, the
percentage agreement was calculated1 and showed sufficiently high scores (the level of agreement for the visual emotions frame was 95% while textual and visual militarism frames, as well as the threat frame, had a 100% agreement between the two coders).
The use of sources. For someone to be counted as a source, a quote, a statement or a fact must have been verbally presented within the news item by anyone other than the reporter covering the story or the anchor of the news (based on Strömbäck et al., 2013; Dimitrova & Strömbäck, 2009b). Only the first three sources appearing in a news item were coded. A set of seven measures was developed based on the PIREDEU project (EES, 2009) and studies of Dimitrova and Connolly-Ahern (2007), Strömbäck et al. (2013), Dimitrova and Strömbäck (2009b), Kim and Weaver (2003). The major components of the sources analysis (for a full codebook and coding instructions see Appendix 1) were the type of source, whether the source is incumbent or opposition (only for Lithuanian, Ukrainian and Russian politicians), the evaluation of the news source in a news story (positive, negative or neutral) and its position (whether the source is neutral, defends or condemns Russia’s position towards the conflict). The inter-coder reliability test yielded sufficient scores for the type of source (α = .88) and the position of the source (α = .84). As there was a low variation in the selected set of test sample of the evaluation of the sources and its power (incumbent vs. opposition), the percentage agreement was calculated, resulting in 100 % agreement for both variables.
All in all, 1593 sources fit the selection criteria and were analysed in the study. Corresponding to the number of news reports, the highest amount of sources was found in PBK (n = 771) while there were less sources in LRT (n = 510) and TV3 (n = 312).
1
Krippendorff’s alpha cannot be calculated for the samples with very low variances therefore, a more
conservative inter-coder reliability test of percentage wise agreement is used. See Vogelgesang and Scharkow, 2012.
Results of content analysis
The analysis of video reports shows that Russian PBK devoted more time for the Ukrainian crisis as compared to Lithuanian media. There were almost 24 hours of coverage in PBK that fit the selection criteria as compared to 10 hours of coverage in LRT and 6 hours in TV3. On average, Lithuanian news reports were 3 minutes long, while in PBK the average length was 5 minutes.
The presence of frames in media coverage
Verbally, the conflict frame (M = .88, SD = .25) and the emotions frame (M = .45, SD = .32) were the most prominently employed frames in the overall news coverage of the Ukrainian crisis (see Appendix 4). The economic frame (M = .18, SD = .28), the diplomacy frame (M = .16, SD = .27), the military frame (M = .13, SD = .30) and the threat frame (M = .04, SD = .18) were less prominent, though they still contribute to the portrayal and
understanding of the overall news coverage of the Ukrainian crisis.
The analysis of visual frames shows similar results. The visual conflict frame (M = .45, SD = .40) and the visual emotions frame (M = .36, SD = .30) also dominated the coverage while the visual militarism frame (M = .27, SD = .39) was less dominant (see Appendix 5).
The results of the prominence of frames within each media outlet are similar: the conflict and emotions frames were the most prominent in textual and visual coverage of the Ukrainian crisis across all outlets, however, to a different extent. Five out of the nine frames analysed in this study differed significantly between Lithuanian media outlets. The
prominence score was higher in privately owned TV3 channel for the textual emotions frame (M = .45, SD = .28), t(321) = 2.40, p <.05, the visual emotions frame (M = .36, SD = .26), t(321) = 2.48, p <.05, the visual conflict frame (M = .63, SD = .41), t(321) = 5.98, p <.05 and the visual militarism frame (M = .36, SD = .41), t(321) = 3.16, p <.05 as compared to the prominence scores M = .37 (SD = .30), M = .29 (SD = .26), M = .36 (SD = .36) and M = .22 (SD = .35) of the respective frames in the Lithuanian public broadcaster. However, the
economic frame was more prominent in LRT (M = .19, SD = .28), t(321) = 2.09, p <.05 than in TV3 (M = .13, SD = .23). As there are significant differences in terms of the prominence of frames, these two Lithuanian outlets will be further compared to the Russian coverage
separately.
Significant differences between Russian PBK and Lithuanian LRT can be seen in the prominence of four frames (see Appendix 6 and Appendix 7). The prominence scores of the textual emotions frame (M = .51, SD = .33), t(478) = 4.65, p <.05 and the visual emotions frame (M = .41, SD = .34), t(478) = 4.59, p <.05 were significantly higher in the news of PBK as compared to M = .37 (SD = .30) and M = .29 (SD = .26) of the respective frames in LRT. However, the latter had significantly higher scores of the conflict frame (M = .91, SD = .23), t(478) = 2.59, p <.05 and diplomacy frame (M = .21, SD = .30), t(478) = 3.53, p <.05 compared to the Russian news (M = .85, SD = .29 and M = .12, SD = .23 respectively).
Five frames were significantly different between PBK and TV3 (see Appendix 6 and Appendix 7). The textual conflict frame (M = .90, SD = .17), t(409) = 2.29, p <.05 and the visual conflict frame (M = .63, SD = .41), t(409) = 4.60, p <.05, as well as the textual militarism frame (M = .19, SD = .36), t(409) = 2.53, p <.05 and the visual militarism frame (M = .36, SD = .41), t(409) = 2.10, p <.05 were more prominent in TV3 compared to M = .85 (SD = .29), M = .43 (SD = .40), M = .10 (SD = .27) and M = .27 (SD = .39) of the respective frames in PBK. The economic frame (M = .19, SD = .30), t(409) = 2.24, p <.05 was more prominent in PBK, than in TV3 (M = .13, SD = .23).
The threat frame developed in this study did not differ significantly across media outlets as determined by an ANOVA test, i.e. its use was similar in terms of prominence across the outlets: F(2,60) = .60, p > .05: M = .05 (SD = .22) in LRT, M = .03 (SD = .16) in TV3 and M = .04 (SD = .16) in PBK.
A comparison of the prominence of frames between the two periods included in the study (1 December 2013 – 17 March 2014 and 10 July 2014 – 7 August 2014) shows that there are significant differences in the dominance of seven frames (see Appendix 8). In the overall coverage of the Ukrainian crisis, the textual emotions frame (M = .66, SD = .75), t(604) = 11.72, p <.05 the textual militarism frame (M = .17, SD = .35), t(605) = 2.06, p <.05, the visual emotions frame (M = .52, SD = .77) t(605) = 8.33, p <.05 and the visual militarism frame (M = .54, SD = .44) t(605) = 11.25, p <.05 were significantly more prominent in the coverage of July and August as compared to the early stages of the crisis (M = .35, SD = .27; M = .11, SD = .27; M = .28, SD = .23 and M = .15, SD = .29 respectively). The textual conflict frame (M = .91, SD = .22), t(605) = 4.55, p <.05, the economic frame (M = .20, SD = .30), t(605) = 3.68, p <.05 and the diplomacy frame (M = .19, SD = .28), t(605) = 3.1, p <.05 were significantly more prominent at the beginning of the conflict and their prominence decreased in July and August (M = .81, SD = .29; M = .12, SD = .22 and M = .12, SD = .24 respectively).
Significant differences in the prominence of certain frames is visible within each media outlet separately as well. In both Lithuanian media outlets the textual emotions frame was more prominent in July and August (LRT: M = .52, SD = .31, t(59) = 3.5, p <.05; TV3: M = .57, SD = .28, t(77) = 3.5, p <.05) compared to the first months of the coverage (LRT: M = .33, SD = .28; TV3: M = .39, SD = .26). Similarly, the visual militarism frame (LRT: M = .56, SD = .42, t(51) = 6.34, p <.05; TV3: M = .67, SD = .39, t(70) = 6.46, p <.05) increased significantly from the early stages of news reports (LRT: M = .13, SD = .27; TV3: M = .21, SD = .33). The textual conflict frame was significantly less prominent in July and August (LRT: M = .85, SD = .21, t(70) = 2.12, p <.05; TV3: M = .86, SD = .18, t(75) = 2.05, p <.05) compared to the first period of the study (LRT: M = .93, SD = .23; TV3: M = .93, SD = .17). In addition, the textual militarism frame was significantly more prominent in July and August
in TV3 news (M = .31, SD = .45, t(58) = 2.34, p <.05) than in the first months included in this study (M = .13, SD = .29).
There were more changes of the frames over time in the Russian news. The textual emotions frame (M = .74, SD = .29, t(219) = 11.22, p <.05), the visual emotions frame (M = .62, SD = .36, t(174) = 9.01, p <.05) and the visual militarism frame (M = .48, SD = .45, t(167) = 7.13, p <.05) were significantly more prominent in July and August compared to the time period between December and March (M = .36, SD = .26; M = .28, SD = .24 and M = .14, SD = .28 respectively). The textual conflict frame (M = .78, SD = .35), t(178) = 3.35, p <.05, the economic frame (M = .10, SD = .21), t(281) = 4.75, p <.05, the diplomacy frame (M = .06, SD = .15), t(281) = 4.63, p <.05 and the visual conflict frame (M = .29, SD = .29), t(282) = 5.36, p <.05 were significantly less prominent in the second period of the study than in the first one (M = .90, SD = .24; M = .25, SD = .37; M = .17, SD = .25 and M = .52, SD = .44 respectively).
The analysis shows that in the beginning of the crisis the coverage in LRT and PBK did not differ significantly, except for the visual conflict frame which was more prominent in the Russian news (M = .52, SD = .44), t(324) = 4.19, p <.05 than in the Lithuanian public broadcaster (M = .34, SD = .36). However, four frames in the coverage of the second time period of the analysis differed significantly between the two media outlets (see Appendix 9). The textual emotions frame (M = .74, SD = .29), t(151) = 3.91, p <.05 and the visual emotion frame (M = .62, SD = .36), t(152) = 3.76, p <.05 were more prominent in PBK as compared to LRT (M = .52, SD = .32 and M = .38, SD = .36 respectively). However, the latter had
significantly higher prominence scores of the diplomacy frame (M = .25, SD = .33), t(152) = 3.75, p <.05 and the visual conflict frame (M = .46, SD = .38), t(152) = 2.64, p <.05
A comparison between TV3 and PBK show that only two frames differed significantly in the coverage of the first months of the crisis. The economic frame was more prominent in PBK (M = .25, SD = .34), t(255) = 3.53, p <.05 compared to TV3 (M = .12, SD = .24) while the visual emotions frame was more present in TV3 (M = .34, SD = .28), t(255) = 2.13, p <.05 than in PBK (M = .28, SD = .24). The analysis of the second period of the study shows that there were significant differences in six frames present in PBK and TV3. The textual
emotions frame (M = .74, SD = .29), t(151) = 3.27, p <.05 and the visual emotions frame (M = .62, SD = .36), t(152) = 3.50, p <.05 were significantly more prominent in PBK as
compared to M = .57 (SD = .28) and M = .41 (SD = .34) scores in TV3. The diplomacy frame (M = .15, SD = .26), t(152) = 2.23, p <.05, the textual militarism frame (M = .31, SD = .45), t(152) = 2.63, p <.05, the visual militarism frame (M = .67, SD = .39), t(152) = 2.55, p <.05 and the visual conflict frame (M = .67, SD = .42), t(152) = 5.38, p <.05 were more prominent in TV3 in comparison with the scores (M = .06, SD = .15; M = .11, SD = .29; M = .48, SD = .45 and M = .29, SD = .29) of respective frames in the Russian news.
A comparison of visual and textual frames
As mentioned earlier, all visual frames ranked in the same order as textual frames in terms of their prominence. In addition, there was a strong positive correlation between visual and textual frames of emotions (r = .85, p < .01), a moderate correlation between visual and textual frames of conflict (r = .31, p < .01) and similarly moderate correlation between visual and textual militarism frames (r = .30, p < .01). Correlations between the same variables were established when analysing each media outlet separately, though to a different degree: the correlation between visual and textual emotions frames was the strongest in PBK (r = .90, p < .01 as compared to r = .79, p < .01 in LRT and r = .78, p < .01 in TV3), the correlation
between visual and textual militarism frames was the strongest in TV3 (r = .51, p < .01 as compared to r = .27, p < .01 in LRT and r = .17, p < .01 in PBK) and the correlation between
visual and textual conflict frames was the lowest in LRT (r = .21, p < .01 as compared to r = .36, p < .01 in TV3 and r = .38, p < .01 in PBK). Following such results, it can be assumed that there is an association between visual and textual frames in the coverage of the Ukrainian crisis, albeit the extent of this association also depends on the frame.
The use of sources in the coverage
The results of the use of sources provide support for the first hypothesis as a combination of all official sources groups (politicians, government officials and official military personnel) dominated the coverage in all media outlets. 52.4% of sources were officials in LRT, 44.7% in TV3 and 39.2% in PBK.
Looking at each group of official sources separately, politicians were the most prominent sources in the overall coverage. In addition, ordinary citizens dominated the coverage to a similar extent. The latter were most prominent in TV3 (38.8% of total sources used) and PBK (37.2%), and less dominant in LRT (31%) (see Appendix 10). Politician sources were the most dominant in LRT (45.6%), while TV3 (36.1%) and PBK (30.6%) relied on them less often. Lithuanian media outlets also used official military sources (5.5% for LRT and 5.1% for TV3), though the third most dominant type of source in PBK was experts
(9.3%). When referring to military sources, Russian news had more unofficial military sources (4.8%) as compared to official sources (3.5%), while unofficial sources in the Lithuanian media were used to a much lesser extent (1.8% in LRT and 3.8% in TV3).
When discussing politicians as sources, it appears that all three outlets used Ukrainian government politicians the most: 19.4% of the total amount of sources were Ukrainian
politicians in LRT, 15.7% in TV3 and 18.5% in PBK. Lithuanian public broadcaster also often relied on international politicians (12.2%) and Lithuanian politicians (10%). TV3 relied on Lithuanian (7.7%) and international (7.7%) politicians less prominently. PBK used
A comparison of incumbent and opposition politicians as sources shows that all three media outlets relied exclusively on incumbent Russian government politicians, therefore the second hypothesis can be confirmed. There was a somewhat similar balance between
Ukrainian incumbent and Ukrainian opposition government sources: PBK and TV3 relied more on incumbent politicians (58.8% incumbent versus 40.5% opposition for PBK and 55.1% versus 44.9% for TV3) while LRT used opposition politicians more often (58.6%) than the ones holding the office (40.4%). However, this study takes into consideration the fact that there was a change in the Ukrainian government during the analysed period. As Yanukovych fled the country on 21 February 2014, the opposition government at the time took over the office and became incumbent government. The analysis of the data for the time period before the changes in the government show different results. Until 21 February, both Lithuanian media outlets used Ukrainian opposition sources more (67%) than the incumbent ones (33%), while incumbent Ukrainian politicians were more dominant (73.6%) in the Russian news compared to the opposition ones (26.4%).
In addition, a chi-square test was used in order to assess the relationship between the power of the Ukrainian politician (incumbent or opposition) and the evaluation of the source in a news story. The results show that a statistically significant association between these two variables is established only in PBK (χ²(1) = 33.60, p = .000) and the level of association is high (.804). All the incumbent Ukrainian politicians were evaluated neutrally or not evaluated at all, while opposition Ukrainian sources were evaluated mostly negatively (71.4%) or evaluated neutrally/ not evaluated at all (28.6%).
The balance of pro-Russia and anti-Russia sources among outlets is mixed. 54.5% of sources in LRT were negative towards Russia as opposed to 15.2% being positive and the rest not expressing evaluation / being neutral. TV3 also relied more on anti-Russia sources, though to a lesser extent (33.8% and 16.4% respectively). PBK had the opposite balance as it relied
more heavily on pro-Russia sources (64.4%) as compared to anti-Russia ones (11.7%) and such findings support the third hypothesis.
A chi-square test was used in order to assess the relationship between the position of the source (pro/anti-Russia) and the evaluation of the source in the news story. The results of the test show that there is a statistically significant association between these two variables in each media outlet (LRT χ²(2) = 46.69, p = .000; TV3 χ²(2) = 45.80, p = .000; PBK χ²(9) = 259.28, p = .000). However, the association is moderate in all outlets (the effect size for LRT is .307, for TV3 .389 and for PBK .338).
Sources that were negative towards Russia did not receive any evaluation or were evaluated neutrally in both Lithuanian TV channels. All anti-Russia sources (100%) received only negative evaluation in the Russian news. The negative evaluation of pro-Russia sources was the most prominent in TV3 news (42.9%) with the rest being evaluated neutrally or not evaluated at all, while in LRT the prominence of negative evaluation was lower (15.4%). In contrast, pro-Russia sources in Russian TV received positive evaluation (22.2%) while the rest were evaluated neutrally or not evaluated at all.
Discussion
This research provides empirical evidence of the differences in the coverage of the Ukrainian crisis in Lithuanian and Russian television news broadcast in Lithuania. The data from the chosen time period of 1 December 2013 – 17 March 2014 and 10 July 2014 – 7 August 2014 reveal the differences of the coverage in a more nuanced way. Framing theory was employed in the content analysis of television reports and the importance of visual framing analysis was emphasized. In addition, the examination of the use of sources
contributed to the overall findings which show that Lithuanian audiences were exposed to a different coverage of the Ukrainian crisis during the analysed period of time.
The analysis of frames shows that all three media outlets used the conflict and emotions frames most prominently in their overall visual and textual representation of the Ukrainian crisis. However, there was some variation in the prominence of each of these frames across outlets. Russian news relied more heavily on the visual and textual emotions frames as compared to the Lithuanian public broadcaster LRT, though to a similar extent as the Lithuanian commercial outlet TV3. Lithuanian media presented the issue using a conflict frame more often than the Russian media. In addition, PBK and LRT emphasized the
economic frame more often than TV3. The threat frame developed for this study was not very prominent and did not differ significantly across outlets. This opposes previous research suggesting that conflict is often presented using threatening coverage (Gadarian, 2010; Gadarian, 2014), though this particular frame still contributes to the findings of the study.
Even though the analysis of the overall coverage did not show much variation between Lithuanian and Russian news, the examination of two different time periods included in this study provides a more nuanced overview of the differences in the coverage. Data shows that there were only a few significant differences in the coverage across channels in the beginning of the crisis, implying that Russian and Lithuanian media reported about the Ukrainian crisis in a similar way. The presentation of events in July and August differed more, showing that PBK covered the issue differently from Lithuanian media outlets. The prominence of frames in both Lithuanian media outlets differed to a limited extent between the two times periods, though the Russian coverage changed over time to a much greater extent. Previous research showed that the conflict coverage in the Russian media reflects the official stance of the government and country’s foreign policy (Baysha, 2010; Heirich & Tanaev, 2009; Eckler & Kalyango, 2012). Therefore, it can be assumed that significant changes of frames in the Russian news might be related to the fact that the country changed its position towards the Ukrainian crisis by getting involved in the conflict at later stages (e.g. by annexing Crimea at
the end of March 2014 or supporting pro-Russian fighters in eastern Ukraine in July and August 2014).
In addition, framing analysis shows that textual and visual emotions frames were more present in all three media outlets in the second period of the study, though the highest growth could be also seen in the Russian news. The diplomacy frame suggesting that the conflict should be solved with international cooperation and negotiations remained similarly prominent in Lithuanian news and dropped sharply in the Russian news. Finally, while the visual conflict frame increased in Lithuanian news, it decreased in the Russian news. Such results show that in July and August the majority of Russian reports emphasized verbal and visual emotions frames while Lithuanian news presented the coverage with a balance of emotions and conflict frames.
The implications of such findings lie in previous framing research, suggesting that exclusively emotional war presentation in the media evokes certain emotions among the audience which results in greater effects on people’s attitudes and opinions (Powel,
Boomgaarden, De Swert & de Vreese, 2015; Graber, 1996; Coleman, 2010, Gadarian, 2014). Therefore, a sharp increase of the emotions frame in the Russian news might lend stronger effects on the audience exposed to such coverage and has bigger potential to shape people’s views and attitudes towards the conflict.
A comparison of the visual and textual presentation suggests that the results are in line with previous research showing that the visual information reflects what is being presented verbally (Mellese & Müller, 2012; Gamson, 1989; Griffin, 2004). The prominence scores and the significant correlation in all three media outlets imply that there were no differences in the imagery and textual information in the coverage of the Ukrainian crisis. In addition, the factor analysis showed that not all measures of visual frames clustered in the same way as it was initially set up, therefore more research is needed in order to establish a better
operationalisation for visual content analysis. However, the current study suggests that certain frame measures might indeed be understood as universal and applicable for both textual and visual analysis.
The overall analysis of the use of sources and the frames show that there are differences in the coverage of the Ukrainian crisis not only between the countries, but also between different types of media. Even though the use of sources was somewhat similar in both Lithuanian media outlets, the overall use of frames differed significantly. This could be a result of one Lithuanian media outlet being commercial and the other a public broadcaster. However, the reasons behind such differences are outside of the current study’s scope, therefore more analysis should be done in order to provide some evidence.
The results of the use of sources are in line with previous research (Dimitrova & Connolly-Ahern, 2007; Strömbäck et al., 2013; Lawrence, 2000; Manning, 2001; Dimitrova & Strömbäck, 2009a; McQuail, 2010) and provide supporting evidence to confirm the first hypothesis. The official sources dominated the coverage of the Ukrainian crisis, however, citizen sources were also very prominent. Some of the previous studies suggested that
geographical obstacles in war journalism are important in terms of the accessibility of sources (Carpenter, 2007); therefore, the fact that Lithuania and Russia are geographically close to the conflict areas in Ukraine might have contributed to the prominence of citizen sources.
The analysis of the government sources and their power (whether they are incumbent or opposition sources) suggests that the second hypothesis can also be confirmed as Russian media relied only on incumbent Russian government sources. However, the same finding is evident in the Lithuanian news, suggesting that no opposing government voices from Russia can be found in the analysed media.
In addition, PBK also relied on incumbent Ukrainian government politicians more than opposition ones, while Lithuanian media used opposition sources much more often than
incumbent ones. Furthermore, in the cases when opposition sources were used in the Russian coverage, they often received a negative evaluation as compared to the incumbent Ukrainian politicians, who were evaluated neutrally or not evaluated at all.
Another difference in terms of the use of sources is that Lithuanian media used more anti-Russia sources compared to pro-Russia sources. This was particularly evident in LRT. The opposite results are seen in PBK, where most of the sources defended, justified or agreed with Russia’s position or actions towards the Ukrainian crisis. Furthermore, all anti-Russia sources in PBK received only negative evaluation and such results provide support for the third hypothesis. Pro-Russia sources dominated the coverage of the Ukrainian crisis in PBK and even though there were anti-Russia sources present in the news reports, they all received only negative evaluation in the news stories which suggests that there was no space for alternative viewpoints in the Russian media. This reflects the previous research (Baysha, 2010; Heirich & Tanaev, 2009; Eckler & Kalyango, 2012; Teper, 2015) and shows that Russian media does not objectively present opinions and views that are different from the official government’s position and that the coverage of the Ukrainian crisis in the Russian media was one sided.
The results of the analysis also confirm the findings of previous research which showed that Russian media devoted a significant amount of coverage for the Ukrainian crisis (Teper, 2015). A comparison of the total amount of coverage shows that there was more coverage in terms of a number of reports and their length in Russian TV, though the fact that the chosen Russian newscasts are longer than the ones in Lithuanian TV should be taken into consideration when interpreting these findings.
The current study was one of very few conducting a comparative media content analysis of the Ukrainian crisis and it is evident that further research is necessary. There were several limitations that are acknowledged in this study. Firstly, available research on the
coverage of the Ukrainian crisis in Russian media has shown that they used certain words and phrases that resonated the cultural and societal context of Russian audience and in this way evoked certain interpretation of the event (Teper, 2015). The current study acknowledges that such findings could not be revealed by conducting a deductive analysis as examination of such content should be based on inductive content analysis which could reveal more issue-specific frames. Secondly, the Krippendorff’s alpha coefficient could not be calculated for some items due to a low variance in a test sample, therefore a less reliable method of calculating a percentage of agreement was used. Furthermore, some items of the frames received relatively low inter-coder reliability scores, therefore this should be taken into account when interpreting the findings. In addition, as the scale reliability coefficient of the diplomacy frame was quite low, the results of this value should be interpreted with more care, and further research is needed in order to have sufficient confidence in this frame. Finally, the threat frame developed for this study was not as prominent as could be expected following previous research (Gadarian, 2010; Gadarian, 2014). Even though the items of this frame were factorable and had sufficient scale reliability scores, further research is suggested by including more items and adjusting the operationalisation.
Despite several limitations, the current study provides more knowledge about the coverage of the Ukrainian crisis and contributes to the existing framing research. The findings show that even though the operationalisation of visual frames can still be improved, the framing theory is rather universal and can be used beyond textual examination. The results of the factor analysis used in this study show that the human interest frame can be expanded by including more items when analysing the coverage of war. Also, the current study developed the militarism frame and the diplomacy frame and they could be applied for other studies focusing on conflict and war.
Based on the findings presented in this study it can be concluded that Lithuanian audiences were exposed to a different coverage of the Ukrainian crisis, especially in the second period of the analysed events. Literature suggests that the Russian government uses Russian media export to post Soviet countries as a soft power tool in order to increase the support towards Russia and its foreign policy in the conflicts (Martišius, 2014; Basilaia, McKeever & Shaw, 2013; Baysha, 2010; Szostek, 2014). As this study provides empirical findings that Russian news were different from the Lithuanian news and that there was a significant shift in the way Russian media covered the crisis (which might be related to the shift in Russian foreign policy), it can be assumed that this has a potential to affect Lithuanian audiences. The major implication of this is that the interpretations of the Ukrainian crisis might differ in Lithuanian society, which thus might result in different public opinions and polarization of views. In addition, this study provides evidence that can be used in the discussions about the Russian media content in Lithuania and their representation of the events. Finally, as this study provides empirical findings showing the main differences in the coverage across Lithuanian and Russian media outlets, further research can be developed in order to analyse how such differences affect the audience.
References
Aday, S.; Cluverius, J.; Livingston, S. (2005). As Goes the Statue, So Goes the War: The Emergence of the Victory Frame in Television Coverage of the Iraq War. Journal of Broadcasting & Electronic Media, 49:3, 314-331. Retrieved from doi: 10.1207/
s15506878jobem4903_4
Anden-Papadopolous, K. (2008). The Abu-Ghraib torture photographs: news frames, visual culture and power of images. Journalism, 9, 15-30. Retrieved from doi:
10.1177/1464884907084337
Balčytienė, A. (2012). Culture as a guide in theoretical explorations of Baltic media. In: D. C. Hallin and P. Mancini (Eds.), Comparing media systems beyond the Western world, 51-71. Cambridge University Press.
Basilaia, E.; McKeever, R.; Shaw, D. (2013). The natural framing of military conflict news. The 2008 war in Georgia in Resonance, Izvestia and The New York Times. Central European Journal of Communication, 1.
Baysha, O. (2010). Is There a Global Public Sphere? Media Framing of the Russia-Georgia Conflict of 2008. Global Media Journal, Volume 10.
Becker, J. (2004). Lessons from Russia: A Neo-Authoritarian Media System. European Journal of Communication, 06/01/2004, 19(2), 139-163. Retrieved from doi: 10.1177/0267323104042908.
Berkowitz, D. A. (1987). TV news sources and news channels: A study in agenda-building. Journalism Quarterly, 64, 508–513.
Boaz, C. (2005). War and Foreign Policy Framing in International Media. Peace Review, 17:4, 349-356. Retrieved from doi: 10.1080/10402650500374603
Boukes, M.; Boomgaarden, H. G.; Moorma, M.; de Vreese, C. H. (2015). Political News with a Personal Touch: How Human Interest Framing Indirectly Affects Policy
Attitudes. Journalism & Mass Communication Quarterly, 92(1), 121–141. Retrieved from doi: 10.1177/1077699014558554
Brader, T. (2006). Campaigning for Hearts and Minds: How Emotional Appeals in Political Ads Work. Chicago: The University of Chicago Press.
Bryman, A. (2012). Social research methods (4th). Oxford, UK: Oxford University Press.
Bucy, E. P.; Grabe, M. E. (2007). Taking Television Seriously: A Sound and Image Bite Analysis of Presidential Campaign Coverage, 1992–2004. Journal of Communication 57, 652–675.
Cacciatore, M. A.; Scheufele, D. A.; Iyengar, S. (2016). The End of Framing as We Know it … and the Future of Media Effects. Mass Communication and Society, 19 (1), 7-23. Retrieved from doi:10.1080/15205436.2015.1068811.
Carpenter, S. (2007). U.S. elite and non-elite newspapers' portrayal of the Iraq war: a comparison of frames and source use. Journalism & Mass Communication Quarterly, 84(4), 4761-776. Retrieved from doi: 10.1177/107769900708400407
Coleman, R.; Banning, S. (2006). Network TV news' affective framing of the presidential candidates: evidence for a second-level agenda-setting effect through visual framing. Journalism and mass Communication Quarterly, 83(2), 313-328.
Coleman, R. (2010). Framing the pictures in our heads. Exploring the framing and agenda-setting effects of visual images. In: D’Angelo, P.; Kuypers, J. M. (Eds.), Doing news framing analysis. Empirical and theoretical perspectives. New York: Routledge.
Cook, T. E. (1998). Governing with the news. The news media as a political institution. Chicago and London: The University of Chicago Press.
Corbu, N.; Iorgoveanu, A. (2012). No consensus on framing? Towards an integrative approach to define frames both as text and visuals. Revista Română de Comunicare şi Relaţii Publice, 3, 91-102.
Delli-Carpini, M. X. (2005). Journalistic frames and ‘Public journalism’. In: Callaghan, K. and Schnell, F. (Eds.), Framing American politics. Pittsburgh: University of Pittsburgh Press.
De Vreese, C. H. (2003). Framing Europe: television news and European integration. Amsterdam: Aksant.
De Vreese, C.H. (2005). News framing: Theory and typology. Information Design Journal + Document Design 13(1), 51-62.
Dimitrova, D. V.; Connolly-Ahern, C. (2007). A Tale of Two Wars: Framing Analysis of Online News Sites in Coalition Countries and the Arab World during the Iraq War.
Howard Journal of Communications, 18:2, 153-168. Retrieved from doi: 10.1080/10646170701309973
Dimitrova, D. V.; Strömbäck, J. (2005). MISSION ACCOMPLISHED? Framing of the Iraq War in the Elite Newspapers in Sweden and the United States. International Communication Gazette, 67(5), 399-417. Retrieved from doi: 10.1177/0016549205056050
Dimitrova, D. V.; Strömbäck, J. (2009a). Look Who’s Talking: Use of Sources in Newspaper Coverage in Sweden and the United States. Journalism Practice, 3(1), 75-91.
Dimitrova, D. V.; Strömbäck, J. (2009b). The Conditionality of Source Use: Comparing Source Use in U.S. and Swedish Television News. Journal of Global Mass Communication, 2(1/2), 149-166.
Eckler, P.; Kalyango, Y. (2012). Cross-national content analysis of the coverage of the 2008 Russia-Georgia conflict. Journal of Media and Communication Studies, 4(3), 35-45. Retrieved from doi: 10.5897/JMCS11.082
Edy, J. A.; Meirick, P. C. (2007). Wanted, Dead or Alive: Media Frames, Frame Adoption, and Support for the War in Afghanistan. Journal of Communication, 57:1, 119– 141. Retrieved from: 10.1111/j.1460-2466.2006.00332.x
EES (2009), European Parliament Election Study 2009, Media Study Data, Advance Release, 31/03/2010, (www.piredeu.eu).
Enikolopov, R.; Petrova, M.; Zhuravskaya, E. (2011). Media and Political Persuasion: Evidence from Russia. American Economic Review, 101, 3253–3285. Retrieved from doi: 10.1257/aer.101.7.3253
Entman, R. M. (1993) Framing: Toward clarification of a fractured paradigm. Journal of Communication, 43, 4; 51-58. Retrieved from doi: 10.1111/j.1460-2466.1993.tb01304.x
Entman, R.M., J. Matthes, and L. Pellicano. (2008). Framing politics in the news: Nature, sources, and effects. In Wahl-Jensen, K. and Hanitzsch, T. (Eds.) Handbook of journalism studies. Philadelphia: Lawrence Erlbaum.
Esser, F. (2008) Dimensions of Political News Cultures: Sound Bite and Image Bite News in France, Germany, Great Britain, and the United States. The International Journal of Press/Politics, 13(4), 401-428. Retrieved from doi: 10.1177/1940161208323691
Eurobarometer. (2014). Media use in the European Union. Retrieved 11 May, 2016, from: http://ec.europa.eu/public_opinion/archives/eb/eb82/eb82_media_en.pdf
Fahmy, S. (2010). Contrasting visual frames of our times: A framing analysis of English- and Arabic-language press coverage of war and terrorism. The International Communication Gazette, 72(8), 695–717. Retrieved from doi: 10.1177/1748048510380801
