Sage Journals: Discover world-class research

Abstract

When people join in moments of mass protest, what role do different media sources play in their mobilization? Do the same media sources align with positive views of mass mobilizations among the public in their aftermath? And, what is the relationship between media consumption patterns and believing disinformation about protest events? Addressing these questions helps us to better understand not only what brings crowds onto the streets, but also what shapes perceptions of, and disinformation about mass mobilization among the wider population. Employing original data from a nationally representative panel survey in Ukraine (Hale, Colton, Onuch, & Kravets, 2014) conducted shortly after the 2013–2014 EuroMaidan mobilization, we examine patterns of media consumption among both participants and non-participants, as well as protest supporters and non-supporters. We also explore variation in media consumption among those who believe and reject disinformation about the EuroMaidan. We test hypotheses, prominent in current protest literature, related to the influence of “new” (social media and online news) and “old” media (television) on protest behavior and attitudes. Making use of the significance of 2014 Ukraine as a testing ground for Russian disinformation tactics, we also specifically test for consumption of Russian-owned television. Our findings indicate that frequent consumption of “old” media, specifically Russian-owned television, is significantly associated with both mobilization in and positive perceptions of protest and is a better predictor of believing “fake news” than consuming “new” media sources.

Keywords

social media television Ukraine EuroMaidan protest disinformation fake news participation

Introduction

When people join moments of mass protest, what role do different “old”¹ and “new”² media sources play in their mobilization? Is the consumption of specific media sources also associated with certain views of the protests among the general public? If so, is there a relationship between “new” and “old” media consumption patterns and believing disinformation about the protests? With increasing numbers of mass mobilizations taking place around the globe, and accusations about the role of “fake news” in driving these mobilizations abound, addressing these questions will help us to better understand not only what brings crowds onto the streets, but also what shapes attitudes toward them and the likelihood of accepting disinformation about mass mobilization among the wider population.

While we know that “new” media have been used extensively during mass mobilizations such as the EuroMaidan, Occupy, and the Arab Spring (Jost et al., 2018; Khamis et al., 2012; Trottier & Fuchs, 2014; Tufekci, 2017), there is a lack of consensus in the literature about whether “new” media actually mobilizes protest participants (Garrett, 2006; Onuch, 2015a; Segerberg & Bennett, 2011). Uncertainty also remains about the effects of social media consumption in shaping public dispositions or driving the acceptance of disinformation campaigns more broadly (Allcott & Gentzkow, 2017; Spohr, 2017; Zhukov, 2017). Past studies have shown that “old” media is significant for both the public’s engagement in and opinion of contentious episodes (Iyengar, 1987; Prior, 2006; Snow & Benford, 1988). Yet, much attention is still paid to “new” media as not only the main source behind mobilization, but also the tool most often used by external powers to manipulate public perceptions of protesters.

To explore contradictions relating to the role of “old” and “new” media in mobilization and mass perceptions of protest events we employ an original, nationally representative panel survey covering participation in and public opinion of Ukraine’s 2013–2014 EuroMaidan mass mobilization, as well as online and offline media consumption (Hale et al., 2014). We analyze whether individuals’ media consumption patterns correlate with participation in, perceptions of, and belief in disinformation about the protests. We also control for a variety of ethnolinguistic, socio-economic, and demographic factors, enabling us to test alternative explanations such as “affective polarization.”

Our findings indicate that frequent consumption of “old” media is strongly associated with protest mobilization. In addition, we find that “old” media consumption is a predictor of whether an individual holds positive views of and believes “fake news” about the protests depending on the “old” media outlet they consume. Specifically, we find robust evidence that frequent consumption of political news on Russian-owned television channels (henceforth, Russian television³) increases the likelihood that respondents also believed disinformation about the EuroMaidan. Our findings not only highlight that “old” media remains important for public opinion, but also underline Russian television’s influence in Ukraine in inflaming tensions and furthering political polarization.

Given the nature of observational data, our article does not provide direct causal evidence and we should be careful not to overstate the influence of either type of media as directly shaping attitudes or behaviors. Yet, even given legitimate concerns about endogeneity, we argue that it is unlikely that people would first arrive at a specific theory about the EuroMaidan and only then seek out a media source to confirm it. Furthermore, we identify strong and robust correlations (or their absence in the case of “new media”) which are in themselves important findings. Our negative finding on the relationship between “new” media and protest participation is counter-intuitive to current common wisdom, thus suggesting limitations on new forms of public influence through “new” media.

In the sections that follow, we first outline the case of Ukraine’s 2013–2014 mass mobilization. We then develop theoretical expectations for our study, based on the existing literature on protest mobilization, public opinion, and media consumption. After this, we describe our data, and methodological and analytical approaches. We finish by discussing our findings, their limitations, and highlighting avenues for further research.

EuroMaidan: Overview

On 21 November 2013, a few thousand Ukrainians took to the streets in major cities, responding to calls expressing outrage at then-President Viktor Yanukovych’s refusal to sign a set of free trade and association agreements with the European Union (Onuch & Sasse, 2016). That night, a new hashtag was born (#EuroMaidan) which gave a name to the emerging protest wave. An “official” “ЄвроМайдан” (EuroMaidan) Facebook page became Ukraine’s fastest growing page, attracting more than 100,000 followers by 1 December. The early, and highly visible role of social media lead some observers to dub the EuroMaidan protests a “Hashtag”, “Facebook” or “Social media” Revolution (BBC Trending, 2013; Hilleary, 2014; Vlasov & Leonard, 2014). And on 24 November, activists and opposition leaders coordinated a march drawing 200,000 participants to Kyiv.

The protests reached a critical turning point when protesters in Kyiv were brutally beaten by “Berkut” militia forces on the night of 29–30 November. Images of the repression were tweeted, posted, and even streamed, triggering widespread outrage. As hashtag use swelled—from 3,200 tweets per hour following the 24 November march to 4,800 per hour on 30 November (Lokot, 2013; Moroz, 2013)—so did protester numbers.

On 1 December, an estimated 2 million people marched in Kyiv and thousands more joined in the events across all of Ukraine (Onuch & Sasse, 2016). Protester demands expanded to universalist claims—the protection of basic human rights and democracy, and the punishment of corrupt elites. Most importantly, the protesters demanded the President’s resignation (Onuch, 2014). A cross-cleavage coalition of individuals from different backgrounds developed, surprising both the party in power and the opposition (Onuch & Sasse, 2016).

The use of “new” media was not limited to social media—Internet-based streaming sites such as SpilnoTV, HromadskeTV, and UkrStream were also important. Near nonstop broadcasts made it possible for anyone in Ukraine or around the world to follow live. “New” media seemed to rule supreme. However, the livestreams were not used by those standing on the square but rather by those sitting on the sofa at home⁴ (Stepanchuk, 2014). In fact, during Onuch’s (2015a) onsite protest survey, a majority of protesters said that regular television as well as friends and family remained their two main sources of information on where, when, and how to protest.

The protest lasted for 3 months, with several cycles of government escalation and reactive use of violent repertoires by protesters. Social media remained an important source for coordination, but also had negative impacts. Social movement organizations, like Hromadskyi Sektor (Civic Sector), used Facebook to disseminate instructions on how to prepare for protest or how to spot and stop provocateurs; lists of what the protesters needed; as well as a running tally of casualties and arrests. Yet, these “new” media also enabled the regime to track protesters, exposing them to violent attacks (Onuch, 2015a), and facilitated the spread of false information about organizers and funders (Onuch, forthcoming). Images of participants wearing balaclavas or engaging in disobedience were coupled with statements about how radical, xenophobic, and criminal they were. The proliferation of new digital technologies also facilitated both the editing of images and their instantaneous distribution. And EuroMaidan leaders lacked any real capacity to control the protest’s image or message in this dynamic media environment.

Increasingly, Yanukovych’s regime and Russian interests were seen as coterminous. Observers of Russian “fake news” narratives that began during the EuroMaidan cite two conspiracy theories central to disinformation campaigns both online and offline. The first was that the United States and the European Union (EU) had orchestrated the protests. Referring to the protests as a coup, the supposed goals of the United States and the EU were to destabilize Ukraine and the broader region, while denying Russia its rightful concerns over Ukrainian affairs (StopFake.org, 2014b, 2014c). The second narrative foregrounded radical-right wing groups’ alleged importance, falsely claiming them as both the majority of protesters and the EuroMaidan’s main organizers (Bezruk, 2014; EU vs DISINFORMATION, 2019a, 2019b).

The protests descended into extreme violence with a final climax over 18–20 February, when more than 100 protesters and 13 militias were killed, and thousands were injured in clashes. These deaths were tweeted, posted, shared, and streamed—but so were frightening pictures of gun-toting protesters. Following a protester ultimatum as well as defections by key oligarchs, President Yanukovych fled the country (Onuch & Sasse, 2016). Despite this apparent victory, the larger political crisis in Ukraine had only just begun. In subsequent months, the Russian Federation annexed the Crimean Peninsula, a major violent episode occurred in Odesa,⁵ and a Russian-instigated insurgency broke out in the Donbas⁶ region. Observers felt that social media networks (specifically Russian-owned VKontakte and Odnoklasnyky) continued to play a vital role in shaping these events. It was also believed that Russian television channels were engaging in a massive disinformation campaign working to shape the views of ordinary Ukrainians (Fedor, 2015; Gruzd & Tsyganova, 2015).

The ensuing deadly conflict, which has killed 13,000 and displaced 2 million (Office of the United Nations High Commissioner for Human Rights [OHCHR], 2019), has been accompanied by information warfare promoting disinformation about the EuroMaidan, the Ukrainian state, and the Donbas (Jaitner, 2015; Mejias & Vokuev, 2017). The threat posed by social media and Russian television disinformation was perceived to be so consequential that the Ukrainian government blocked VKontakte and 14 Russian television channels (The Economist, 2017; President of Ukraine, 2014). Meanwhile, EuroMaidan activists set up fact-checking organizations like “Stop Fake,” to combat Russian media disinformation (StopFake.org, 2014a).

But was disinformation around the EuroMaidan really a #Revolution,⁷ mostly benefiting from “new” media proliferation and consumption? Or was Russian television (and its consumption) the main and more successful menace in the campaign to disinform and polarize Ukrainian citizens? And thus, do patterns of media consumption relate to Ukrainians’ participation in, evaluations of, and acceptance of disinformation about the EuroMaidan? To address these questions, we turn to current scholarship on protest, media, and mass perceptions.

Media, Mobilization, and Mass Perceptions: Framing the Analysis

Media Consumption and Political Participation

What is the relationship between media consumption and protest participation? Like all studies on the relationship between media consumption and political behavior, the scholarly literature is divided (Clarke & Fredin, 1978; Glaser, 1965; Hayes, 2009; Stroud, 2008). While Gentzkow (2006) found that television diminished political engagement, Sørensen (2019) found that consumption of public broadcasting TV can increase levels of political participation, suggesting that specific channel content and consumption mattered. Furthermore, scholars note that without conducting randomized control trials or “natural” experiments (Peisakhin & Rozenas, 2018), it is difficult to make causal claims about media consumption driving political behavior due to possible self-selection bias and endogeneity (Salzman & Aloisi, 2009), which can muddle inferences based on observational data.

Recent scholarship on the role of the Internet and social media maintains similar divisions. While Tolbert and McNeal (2003) found that respondents accessing election news online were significantly more likely to report voting, their socio-economic status and education were highly likely to systematically influence this access. Wellman et al. (2001) found a positive association between offline and online participation in politics, yet others highlight that social media can depress offline political engagement as individuals achieve satisfaction from “clicktivism” (Jones & Wayland, 2013). Most of these studies concede that when standard control variables such as age, education, race, socio-economic class, and urban residence are accounted for, much of the effect of social media or online news use is mediated (Enikolopov et al., 2011; Wei & Hindman, 2011). This research highlights that different types of individuals will use the Internet differently (Vitak et al., 2010). More specifically, it suggests that the usual suspects of the politically aware and socio-economically well-off are more likely to read and share political news online—although the less politically engaged also do so (Vitak et al., 2010).

Media Consumption and Protest Participation

It is no surprise that recent studies examining media consumption patterns and protest behavior focus on “new” media. Research on mass mobilization has repeatedly highlighted protesters’ usage of “new” media, such as online news sites, blogs, and particularly social media (Khamis et al., 2012; Trottier & Fuchs, 2014). “New” media has been identified as both a new tool for activists to coordinate activity, and as a means for mass communication with or between ordinary citizens. The speed and low-cost with which “new” media transmits information about protest facilitates its mobilizing potential (Bennett, 2013; Segerberg & Bennett, 2011; Youmans & York, 2012). “New” media is considered particularly powerful because of its capacity to reach across multiple diverse networks, bridging weak ties, and thus, “scaling” up protests (Tufekci, 2017; Tufekci & Wilson, 2012)—but, this easy-entry model also leaves protests with unstable, momentary, and virtual coalitions.

There remains little consensus about the role “new” media play in the motivation and mobilization of protesters (Garrett, 2006; Segerberg & Bennett, 2011). Studies by Jost et al. (2018) and Onuch (2015a, 2015b) question social media’s power as a tool of mobilization, highlighting that while it is very likely that “new” media may help motivate some individuals to consider joining a protest event, they are unlikely to push people on to the streets per se. This last step requires participants to be embedded in offline personal networks with high levels of trust, which in turn limit the risks associated with participation. These ties are far less likely to be present in solely online networks. Several studies furthermore suggest that consumption of “old” media, and here, we specifically mean television, continues to be related to the mobilization of “ordinary” citizens during contentious episodes (Beissinger et al., 2015; Boyle & Schmierbach, 2009; Snow & Benford, 1988).

In the Ukrainian context, onsite surveys of EuroMaidan participants show that the top source of information was television, at 58% of respondents (Onuch, 2014). According to a nationally representative survey, 54% of participants learned when or where to go to protest from TV and radio (Hale et al., 2014). This fits with findings that traditional media remain the main information source for most people in contemporary societies and continue to dominate information dissemination to the general public (Papathanassopoulos et al., 2013).

Given all this, we believe “new” media (1) can provide significant capacity gains through coordination mechanisms, (2) can broadcast information quickly and disseminate it broadly among the Internet-using segment of the population, and (3) may motivate and appeal to the politically engaged due to connecting political goals or emotions to protest goals. But, given the Internet’s low-trust environment, as well as the lack of true connections to protesters through social ties, this may not translate to mobilization proper. Furthermore, the continued use of television as a primary news source means that protest information dissemination will reach more people through TV than through “new” media, and specifically social media. Moreover, television is seen by many as more trustworthy and less open to manipulation than online and print media (Newman & Fletcher, 2017; EBU, 2017). Thus, we distinguish the effects of “old” media from “new” media, and propose the first element of our “old” media hypothesis.

We expect,

H1. Protest participation to be positively and significantly associated with television news consumption.

And conversely, we propose the first element of our “new” media hypothesis, and expect that

H2. There will not be a significant association between protest participation itself and “new” media consumption.

But this still leaves the question of whether the same patterns of media consumption are similarly related to, and potentially shape, mass perceptions of protest events?

Media Consumption and Perceptions of Protest

Scholarship on media consumption and public opinion of protest is similarly conflicted. McCombs and Shaw (1972) and McCombs (2014) have argued that “old” media influence mass perceptions of important events by “setting the agenda” and influencing public perceptions of reality. These findings have been extended to studies of mass perceptions of protest. As noted by Crabtree et al. (2015) and McLeod and Detenber (1999), the framing and agenda-setting powers of different television channels have been found to be related to the likelihood that an individual feels solidarity with or even joins a protest. Despite expectations that “old” media’s ability to shape mass perceptions would decline with the rise of the Internet (Chaffee & Metzger, 2001), studies indicate that “old” media continue to influence public opinion during crises (Djerf-Pierre & Shehata, 2017; King et al., 2017).

Meanwhile, scholars disagree on the extent to which “new” media shape mass perceptions more broadly (Murphy et al., 2014; Shapiro & Jacobs, 2011) and of protest specifically (Garrett, 2006; Segerberg & Bennett, 2011). There is some limited evidence that social media content can negatively impact perceptions of protesters (Johnson & LeFebvre, 2018) and that authoritarian states seek to manipulate the public and undermine the opposition with social media (Spaiser et al., 2017). Nonetheless, much of this literature warns that different medias’ effects on public opinion are highly contextual. We, thus, need to understand the specific media landscape in the country under study.

According to studies of the Ukrainian media landscape⁸ (Dyczok & Gaman-Golutvina, 2009; Szostek, 2014b), television remains the dominant news medium, used by 97% of the population (BBG Gallup, 2014). Despite a post-EuroMaidan ban on Russian television in Ukraine, 19% of citizens in south-east Ukraine continued to watch it using a satellite dish, and 38% use a virtual private network (VPN) to stream the broadcast (BBG Gallup, 2014). Going online remains the second-most used news source, at 48%. However, usage varies significantly by age, ranging from 90% for those aged 15–24 to only 12% of those over 55 years (BBG Gallup, 2014).

In terms of social media,⁹ Russian-owned VKontakte and Odnoklasnyky dominated the Ukrainian market in 2014, with 75% and 66% of the population registered (Yandex, 2014). Despite much discussion of the role of Facebook and Twitter during EuroMaidan, only 10% of the population was signed up to Facebook, and under 1% to Twitter by the end of 2014. Considering the Ukrainian media context and findings about the continued influence of television, there is no reason why public perceptions of the protest would be less likely to be linked to “old” media consumption. Thus, we extend our “old” media hypotheses to also postulate that

H3. Public views of protests (positive and negative) would be more closely aligned to and will correlate strongly with the consumption of “old” media as opposed to “new” media.

However, the main issue around media consumption and mass perceptions of EuroMaidan was the suspected role of disinformation campaigns (Dyczok, 2014; Mejias & Vokuev, 2017).

Media Consumption and Buying into Disinformation

Disinformation campaigns are not new. Yet, with the rise of “new” media, scholars have become particularly concerned with social media’s ability to spread false information for political purposes (Howard & Bradshaw, 2018).

Social media is considered a particularly dangerous source of disinformation because (a) platforms are susceptible to the use of state-sponsored bots¹⁰ and trolls¹¹ which aim to magnify marginal, divisive, or conspiratorial opinions (Howard & Bradshaw, 2018); and (b) misleading or fabricated content is often disseminated much more quickly on social media than through traditional media (Wardle, 2017). Nevertheless, disinformation is not a new phenomenon and was disseminated in newspapers and on television long before the Internet emerged (Chesney & Citron, 2019). Even today, traditional media in the form of state-run, elite-backed, or partisan TV channels attempt to shape public narratives through sharing inaccurate, misleading, or false information (Benkler, 2018; Walker & Orttung, 2014).

In the case of Ukraine, Mejias and Vokuev (2017) have noted that both Russian television and social media¹² have been used to spread disinformation. Scholars have highlighted that the content of Russian television and the Russian social media sites (namely, VKontakte and Odnoklasnyky) were less supportive of the EuroMaidan, European integration, and Ukrainian independence (Dyczok, 2014; Kozachenko, 2014; Kulyk, 2014; Szostek, 2014a). This leads us to propose our Russian media hypotheses.

First, following on from our above thinking on the role of “old” media, and the literature on the Ukrainian media context, we propose our Russian television hypothesis, whereby we expect that

H4a. Consumption of Russian television will be negatively correlated with holding positive views of the EuroMaidan and will correlate positively with buying into disinformation.

And if we were to base our thinking on this literature focused on the Ukrainian media context, we would expect the same to be true in the case of consumption of Russian “new” media. Namely that

H4b. Individuals who use Russian social media (VKontakte and Odnoklasnyky) will be less likely to hold positive views of the EuroMaidan and more likely to buy into disinformation about it.

Yet, we expect that consumption of “old” as opposed to “new” media will align with general public views of protests and thus propose a competing hypothesis to the earlier. In line with both Nisbet and Kamenchuk (2019) and Lewandowsky et al. (2017), we redirect the focus away from technological innovations in social media as driving disinformation. We decouple our treatment of how disinformation travels and is consumed in these two media outlets (Russian television and Russian social media). Instead, we argue for an approach that privileges the cognitive responses of actors embedded within society—whereby, underlying partisanship, identity-based orientations and worldviews, and socio-economic characteristics allow disinformation not only to spread but also, and crucially, be accepted.

Thus, we highlight two mechanisms through which individuals’ beliefs and identities can influence their susceptibility to disinformation: “motivated reasoning”¹³ and “affective polarisation” (Nisbet & Kamenchuk, 2019). Processes of “motivated reasoning” can lead individuals to seek out and accept information compatible with their beliefs, thus making them more susceptible to disinformation which appeals to their partisan position (Flynn et al., 2017; Lodge & Taber, 2013).

This process may be amplified by social media: not only do social media users have to first seek out certain information and then “like” or share it, but social media algorithms can work to exacerbate this selection mechanism and self-placement into neatly partisan or “siloed” information bubbles (Pariser, 2011; Stroud, 2010). However, some research has challenged the existence of these “filter bubbles” (Nelson & Webster, 2017), indicating that social media also exposes individuals to material which challenges their partisan beliefs (Barberá et al., 2015), particularly when their online networks comprise of diverse individuals (Bakshy et al., 2015; Flaxman et al., 2016).

Moreover, and particularly important for our thinking, this process may play out differently among users despite using identical platforms. Two people using the same social media site, at the same time and day, in the same location, could have two very different experiences of the information supplied to them—precisely because of self-selection processes, as well as actual reasons for using the platform.¹⁴ It is for this reason that we would not expect social media use to be significantly associated with a particular type of disinformation. Therefore, if our thinking (competing with the above [H4b], as well as popular and government discourse around perceptions of key channels of disinformation in Ukraine) is correct, we can expect that

H4c. Users of Russia social media (VKontakte and Odnoklasnyky) will be no more likely to buy into disinformation about the EuroMaidan or to have negative or positive views of it.

Meanwhile, we postulate that because two people exposed to the same television channel receive the same information simultaneously, the only thing someone who strongly disagrees with the content can do is to stop watching this channel. Thus, we propose our frequency of viewership hypothesis, expecting that

H5. Watching Russian television channels frequently would be associated with the likelihood of holding negative views of the EuroMaidan and agreeing with disinformation.

While we place more emphasis on the role of media, we acknowledge that disinformation targets particular groups. Thus, our thinking about the mechanism explaining believing disinformation about the EuroMaidan is closer to the second mechanism of “affective polarisation.” Affective polarisation occurs when political discourses and framings increase an individual’s positivity toward those perceived to share their identity, and hostility toward “outsiders” (Wojcieszak & Garrett, 2018). In the context of key Russian disinformation narratives about the EuroMaidan, this is visible in its focus on both the malign intent of foreign actors (EU, the United States or CIA) in orchestrating the protest, but also on the Ukrainian-ethnocentric and violent nature of the protests—the target of which were eastern Ukrainians, ethnic Russians, and Russophones.

There is consensus that overlapping regional, ethnic, and linguistic divides drive political participation and public opinion in Ukraine (Arel, 1995; Barrington & Faranda, 2009; Barrington & Herron, 2004; Kulyk, 2001; Laitin, 1998). More recently, scholars, such as Kulyk (2019) and Onuch and Hale (2018), caution against an overly simplistic view of ethnicity and language as shaping Ukrainian political behavior. Onuch and Hale (2018) advise that while region is quite a powerful predictor, language practice, language embeddedness, ethnolinguistic identity, and national identity all measure different things and have divergent effects on behavior and opinion.

Russian disinformation campaigns about a foreign-organized and funded coup referenced a need to defend “Russian speakers” from the likely onslaught of Ukrainian nationalists (Bezruk, 2014; EU vs DISINFORMATION, 2019a, 2019b; StopFake.org, 2014b, 2014c). Given the longstanding ethnolinguistic cleavage in Ukrainian politics, and associated dislike among many Russophones and residents of south-eastern Ukraine for Ukrainianization (thought to be supported by foreign governments), narratives that fed into pre-existing perceptions were plausibly more likely to be believed through “affective polarization.” We expect that residence in eastern and southern regions make it more likely that individuals see the EuroMaidan as something external to their identity, increasing their hostility toward these “outsiders,” as well as their likelihood of believing disinformation about them. Conversely, those who regularly speak Ukrainian, who are embedded in Ukrainian-speaking work environments, have a strong sense of Ukrainian ethnic identity, and those who have strong sense of Ukrainian civic or national identity should be less likely to connect to disinformation narratives.

For this reason, we propose the first two elements of a three-part affective polarization hypothesis. We expect that

H6. Ukrainophones, ethnic Ukrainians, western Ukrainians, and those with a strong sense of Ukrainian civic identity, would likely identify with the protests and would be unlikely to believe the content of disinformation narratives.

and

H7. Russophones, ethnic Russians, eastern and southern Ukrainians, and those without a strong sense of Ukrainian civic identity will be more likely to accept disinformation narratives about the EuroMaidan.

Recent studies have also repeatedly shown that those who were economically vulnerable, or perceive themselves to be economic losers are also more likely to support Russian-backed separatism and believe disinformation about the conflict (Giuliano, 2018; Zhukov, 2016). Russian disinformation narratives also often referenced a Ukrainian economic crisis, harked back to past economic successes of the communist era, and made references to a Kyiv elite that did not care about poor or working people (StopFake.org, 2014c). Keeping in mind that buying into disinformation may be associated with background factors, it is also plausible that disinformation narratives that fed into pre-existing perceptions of Kyiv “elites” can be believed more easily through affective polarization. And thus, in line with our affective polarization hypothesis, we also expect that

H8. Those who perceive themselves to be economic “losers” will be more likely to accept disinformation narratives about the EuroMaidan.

The inclusion of variables associated with the affective polarization hypothesis provides an important robustness test for our proposed “old” media, Russian television, and frequency of viewership hypotheses. We postulate that if media consumption patterns mediate or absorb some of the effects of belonging to the aforementioned groups then our proposed claims about their role might even be stronger.

Data and Operationalization

To test these theoretical expectations, we make use of an original dataset: the Ukrainian Crisis Election Panel Survey (Hale et al., 2014). This nationally representative three-wave panel survey was conducted in collaboration with The Kyiv International Institute for Sociological Studies in 2014—shortly after the protests concluded. Our analysis below draws upon the first and second waves of this three-wave survey. The first wave comprised of a nationally representative sample of 2015 individuals,¹⁵ and was conducted from 16 to 24 May 2014. The response rate was 51%. The second wave took place 24 June to 13 July 2014, consisting of interviews with 1,406 of the original 2,015 respondents. The margin of error of our frequency estimates is no greater than 3.3%.

Dependent Variables

To enable us to explore whether different patterns of “new” and “old” media consumption can be identified between protest participants and non-participants, we use an item from Wave 1 of the survey. We create a binary variable for EuroMaidan protest participation, taking the value 1 when the respondent reported participating in the EuroMaidan (see Table 1 below).¹⁶

Table 1.

DV1—EuroMaidan Protest Participation (Hale et al., 2014).

	Frequency	Percentage (%)
Not protest participant	1,818	90.22
EuroMaidan protest participant	197	9.78
Total	2,015	100

To test whether those who have positive perceptions of the EuroMaidan hold similar patterns of “new” and “old” media consumption to those who participated in the protest, we employ a Wave 2 survey item whereby respondents were asked to indicate the extent of their positive or negative feelings toward EuroMaidan on a 0–100 scale. Using these responses, we generated a continuous variable indicating the degree of positive perceptions of the EuroMaidan, illustrated below in Table 2.

Table 2.

DV2—Positive Perceptions of the EuroMaidan (Hale et al., 2014).

	Frequency	Percentage (%)	Cumulative percentage (%)
Very negative	190	15.47	15.47
10	89	7.25	22.72
20	71	5.78	28.5
30	73	5.94	34.45
40	48	3.91	38.36
50	232	18.89	57.25
60	76	6.19	63.44
70	107	8.71	72.15
80	138	11.24	83.39
90	86	7	90.39
Very positive	118	9.61	100
Total	1,228	100

Finally, to investigate whether believing disinformation about the EuroMaidan protests aligns with the same patterns of consumption as participation in and positive perceptions of the EuroMaidan, we use a multiple-choice survey item from Wave 1 asking respondents who organized the protests. We consider respondents to believe disinformation campaigns if they selected the option that American or European governments mainly organized the EuroMaidan out of four options available, creating a binary variable for disinformation belief (see Table 3).¹⁷

Table 3.

DV3—Believing Disinformation About the EuroMaidan: EU and the United States Were Main Organizers (Wave 1) (Hale et al., 2014).

	Frequency	Percentage (%)
All others	1,833	90.97
European and American governments were the main organizers	182	9.03
Total	2,015	100

Independent Variables

The theory discussed earlier leads us to test the relationship between three sets of “new” and “old” media consumption variables.

Old Media

In Ukraine television is the dominant “old” media news source, with only a fraction of the population using newspapers and radio (BBG Gallup, 2014). Hence, we focus on television consumption. First, we distinguish between those who do and do not watch television (for full Independent Variable survey item details and coding please see Supplemental Appendix Table A4). Moreover, since we have both empirical and theoretical reasons for expecting Russian television (and again, here we mean Russian-owned) to be more likely to present a negative picture of the protests and to have communicated disinformation, we divide our analysis accordingly. To capture this “old” media consumption, we use two wave one survey items which capture whether a respondent watched political news or shows on Ukrainian-owned channels and/or on Russian-owned channels in the 7 days prior. We create separate binary variables for reported consumption of Ukrainian television¹⁸ and Russian television.¹⁹

New Media

To capture “new” media consumption patterns, we employ five binary variables. First, we distinguish between those who do and do not use the Internet. Second, we specifically identify social media users. Since the literature expects Russian social media sites to be more likely to present a negative picture of the protests and to spread disinformation about them, we disaggregate VKontakte, Odnoklasnyky, and Facebook users. We create variables for these three sites as they dominate social media usage in Ukraine.

As the literature suggests that there is a difference between those who use the Internet for news and those who use it for other things like communicating with family, or entertainment, we include a binary variable capturing those who said that they use the Internet primarily for “getting news and information about current events.”

Frequency of TV Watching

We measure consumption frequency by including six scaled variables to account for consumption frequencies of Russian and Ukrainian television as well as social media and Internet news. Including these variables helps to reduce any signaling effect from respondents self-reporting their media use, and accounts for the cumulative effect of consumption. We use two survey items (Wave 1), where respondents were asked how often in the last 7 days they had watched daily news or political shows and how often they had watched news on the Internet to construct an ordinal variable for frequency.

Control Variables

In addition to our primary independent variables, we control for gender, urban residence, age, education, and socio-economic status. And, as we expect that “affective polarisation” may also be at play in the acceptance of EuroMaidan disinformation, we control for region of residence, language practice, language embeddedness, ethnolinguistic identity, national identity, protest participation, and perceiving oneself as an economic loser (see Supplemental Appendix Table A5 for full details of control variables).²⁰

We create four dummy variables for respondents’ residence in the east, west, south, or center macro-regions. Because major differences are found between the east/south and west macro regions, and the center is seen as a mixed region, we use this as a reference category, excluding it from our analysis.

Given the complexity of measuring identity in Ukraine, following on from Onuch and Hale (2018) we break it down into four components. We measure personal language preference with a variable coding all those who chose to take the survey in Ukrainian. To capture language embeddedness, we created a variable coding all those who report using Ukrainian in their place of work. Ethnolinguistic identity was measured with a variable coding all those reporting Ukrainian as their native language. Finally, for our national identity (also referred to as civic identity) variable, we coded all those who selected that if they had to register one nationality, they would register as Ukrainian.

Following Tucker (2006), we capture whether someone is an economic loser with a binary variable coding all those who reported that their family lost or mostly lost in economic terms as a result of the changes since Ukraine’s transition to independence in 1991 (Wave 1).

All our models incorporate a continuous variable for age, a binary variable for gender (Female = 1), a six-point scaled variable for education, and a binary variable capturing residence in an urban environment over 50,000 residents. To capture socio-economic status, we employ a measure of the respondent’s family financial situation.

Causal Sequencing and Analytical Approach

To estimate the relationship between our variables of interest and our dependent variables, we employ two logistic multivariate regressions (for DV1 and DV3) and one ordinary least squares (OLS) multivariate regression (for DV2). As logit coefficients do not facilitate straightforward interpretation, we report estimated effects of each factor on a given dependent variable in terms of full effects for DV1 and DV3. A full effect is the average marginal effect when all variables are scaled from 0 to 1. Therefore, these results should be interpreted as the average change a factor produces in an individual’s estimated likelihood of becoming a protester, or their likelihood of believing that the United States and the EU organized the EuroMaidan, when one raises any given factor from its minimum to its maximum value while holding all other variables at their actual values. Where our dependent variable is continuous (DV2), we report the marginal effects of each factor on the likelihood of seeing the EuroMaidan as positive.

When estimating effects of each factor, our modeling choices for control variable inclusion follow Campbell (1980) and Colton (2000), whereby we include variables of interest at different “causal” stages into our equation. In Stage 1, we include “observed” demographic factors (region, urban residence, language of questionnaire, level of education, gender, and age) unlikely to be driven by other factors of interest. In Stage 2, we separate out what we consider to be “declared” demographic factors (language embeddedness, ethnolinguistic identity, and national identity) as these are more likely to be influenced by first-stage factors. In Stage 3, we include socio-economic variables (family financial situation, being a “transition loser”) and only in the case of DV2 and DV3, we also include EuroMaidan participation. Next, we include our independent variables capturing “old” and “new” media consumption. In Stage 4, we include “old” media (watching Russian TV channels, Ukrainian TV channels, or not watching TV at all). In Stage 5, we include “new” media (using Facebook, VKontakte, or OdnoKlasnyky, or using the Internet for news).

In a separate model, we account for the frequency of watching Russian or Ukrainian television, of using social media sites Facebook, VKontakte, or OdnoKlasnyky, or using the Internet for news. In this model, to avoid collinearity, we exclude the variables added in Stages 4 and 5. As a robustness check for our findings related to believing disinformation, we ran our model with a different binary DV4 (see Supplemental Appendix Table A9 and Figure 4 for these results).

Results and Discussion

The results from our analysis are definitive and we believe highly robust. We confirm our expectations that media consumption patterns align with both participation in, positive perceptions of, and believing disinformation about the EuroMaidan. We also find that their inclusion mediates the effects of residing in the east and south, in some instances completely absorbing this effect. Furthermore, we find that media consumption patterns mediate language practice, language embeddedness, ethnic identity, and nationality, where they were significant in this first stage.

What is perhaps most interesting is that unlike many of the expectations in the literature on protest, and confirming our “new” media hypotheses (H2, H3, H4c), “new” media consumption patterns are not significantly associated with any of our dependent variables. There is no evidence from our analysis that any social media platform correlates with mobilization or beliefs about the protests. It may be surprising to some observers to know that even usage of the Russian (state-sponsored) VKontakte had no effect. We thus reject the hypothesis found in the literature (H4b) expecting that those “individuals who consume news on Russian social media will be more likely to buy into disinformation about the EuroMaidan.”

We find consistent evidence that “old” media (H1 and H3) and specifically frequently watching Russian television (H5) is associated with both protest participation and holding positive views about the EuroMaidan. Even more convincingly, we find that watching Russian television increases the likelihood that an individual believes disinformation about the protests, confirming hypotheses H4a and H5. This confirms our frequency of viewership hypothesis.

Finally, we find only limited evidence supporting the affective polarization hypothesis, finding that perceiving oneself to be an economic loser (H8), residing specifically in the east of the country, and not holding a strong Ukrainian civic identity (H7) significantly shape the likelihood of believing disinformation about the EuroMaidan. We find the same pattern for participating in, or having positive views of, the EuroMaidan. However, we do not find evidence of ethnic or linguistic identities being associated with participation in, views of, or belief in “fake news” about the EuroMaidan. Below, we interpret results for each dependent variable in turn.

Protest Participation

We report full effects for our dependent variable of EuroMaidan protest participation visually with key independent variables in Figure 1, and with control variables in Table 4. We see no evidence for hypotheses that social media usage correlates with protest participation. We confirm our hypothesis (H4c) that watching Russian television frequently correlates negatively with EuroMaidan protest participation. We find that frequently watching Russian television channels is associated, at a statistically significant level, with an 8% decrease in the likelihood that someone was a protest participant. Due to model limitations, we cannot causally say whether Russian television actively demobilized people, but we can be certain that, controlling for a variety of factors, watching Russian television frequently is significantly correlated to not having participated in the EuroMaidan protests. Interestingly, we also find that non-Internet users (about 44% of our sample and by our estimation about 48% of the population) are 6% less likely to have been a EuroMaidan protest participant. Age and not using the Internet are negatively correlated with protest participation. We note that we do not find a similar correlation between frequent Russian television viewership and age.

Figure 1.

Correlates of Euro Maidan participation.

Table 4.

Full Effects of Variables on Probability of Participating in EuroMaidan (Hale et al., 2014).

	Old mediaStage 4	New mediaStage 5	Media frequencyStage 6
All significant control variables
East	–0.09*	–0.09*	–0.08
South	–0.09*	–0.09*	–0.09*
West	0.07**	0.07**	0.07**
Urban locality	0.07**	0.06*	0.06*
Education	0.12**	0.09*	0.08*
Language at work Ukrainian (language embeddedness)	0.04*	0.04*	0.04*
Family financial situation	0.15**	0.13*	0.12*
Explanatory variables
Watches Ukrainian TV channels	–0.02	–0.02
Watches Russian TV channels	–0.04	–0.05
Does not watch TV	–0.05	–0.06	–0.04
Not user of Internet		–0.04	–0.06*
Odnoklasnyky		–0.01
VKontakte		0.01
Facebook		0.01
Uses Internet for news		0.03
Ukrainian TV frequency			0.01
Russian TV frequency			–0.08*
VKontakte frequency			–0.00
Facebook frequency			0.02
Odnoklasnyky frequency			–0.01
Internet news frequency			0.03
N	2,015	2,015	2,015

Calculated using logit model.

p < .05; **p < .01.

Non-significant variables omitted, see Table A2 in Supplemental Appendix.

Unsurprisingly, the strongest predictors of protest participation are education and family financial situation, increasing the likelihood that someone was protest participant by 13% and 15%, respectively.²¹ We find neither to be correlated with frequent TV consumption. We also find an unsurprising negative effect of residing in the east and south, decreasing the likelihood that an individual was a EuroMaidan protest participant by 14% and 11%, respectively.

Positive Views of the EuroMaidan

We repeat the analysis with our second dependent variable measuring overall views of the EuroMaidan. Our results are presented visually with key independent variables in Figure 2, and with control variables in Table 5. There is no evidence that social media usage is associated with any statistically significant change in views regarding the EuroMaidan. We do find robust evidence that watching Russian television, and being a frequent viewer, are correlated with a decrease in the likelihood that someone regards the EuroMaidan positively by 8% and 9%, respectively (H3, H4a). Conversely, being a viewer of Ukrainian television channels increases the average likelihood that someone viewed the EuroMaidan positively by 14% and 9%, respectively. These figures are statistically significant and highlight a strong possible effect of both Ukrainian and Russian television consumption shaping views of the EuroMaidan.

Figure 2.

Correlates of viewing EuroMaidan as positive (Hale et al., 2014).

Table 5.

Marginal Effects of Variables on Probability of Viewing EuroMaidan as Positive (Hale et al., 2014).

	Demographic variablesStage 1	Declared demographic variablesStage 2	Socio-economic variablesStage 3	EM protest participation variablesStage 4	Old mediaStage 4	New mediaStage 5	Media frequencyStage 6
All significant control variables
East	–28.72**	–23.14**	–19.63**	–19.31**	–15.65*	–15.52*	–16.25*
South	–14.18**	–12.30**	–8.29*	–8.04	–8.12	–8.24*	–8.36
Survey in Ukrainian (Language embeddedness)	7.08*	4.96	5.5	5.39	5.35	5.29	5.13
Age	–7.73*	–5.12	–1.51	–1.17	–3.92	–0.42	–1.77
Female	–3.81*	–3.4	–3.58	–3.39	–3.13	–2.96	–3
Nation forced choice Ukrainian (Ukrainian nationality)		14.66**	10.99**	11.11**	11.12**	10.93**	10.85**
Transition loser			–13.08**	–13.03**	–12.06**	–12.16**	–12.14**
EM protest participant				7.98*	7.80*	7.61*	6.64
Explanatory variables
Watches Ukrainian TV channels					13.62**	14.57**
Watches Russian TV channels					–7.75**	–7.48**
Does not watch TV					10.64	10.76	4.14
Not user of Internet						–2.65	1.03
Odnoklasnyky						–3.34
VKontakte						4.07
Facebook						2.85
Uses Internet for news						–2.81
Ukrainian TV frequency							9.10*
Russian TV frequency							–9.23**
VKontakte frequency							6.01
Facebook frequency							–1.36
Odnoklasnyky frequency							–4.87
Internet news frequency							4.23
N	1,228	1,228	1,228	1,228	1,228	1,228	1,228

95% confidence intervals in brackets. Calculated using OLS model.

p < .05; **p < .01.

Non-significant variables omitted, see Table A2 in Supplemental Appendix.

While being a protest participant is associated with holding positive views of the EuroMaidan, this is not the case when we control for frequency of Ukrainian and Russian television viewership. The effect of being a protest participant is completely absorbed. We also find that residing in the east and feeling that one is a transition loser decreases the likelihood that someone will view the EuroMaidan positively by 29% and 13%, respectively, at statistically significant levels. We find a similarly strong but inversely correlated effect of Ukrainian civic identity increasing the likelihood that someone views the EuroMaidan as positive by 15%. Moreover, we note that being from the south of Ukraine is only sometimes correlated at statistically significant levels with negative feelings about the EuroMaidan, although when it is, it is a substantive result.

Belief in Disinformation about the EuroMaidan

Finally, turning to our third dependent variable measuring beliefs in Russian disinformation narratives, we find no evidence that social media was associated at levels of statistical significance with believing in disinformation narratives. Figure 3 represents our results for key independent variables, and Table 6 presents our results including control variables and all independent variables. This confirms our hypothesis H4c and allows us to reject the competing hypothesis (H4b). Instead, we find strong evidence for the divergent impact of broadcast television on buying into disinformation. Viewing Russian television is statistically significantly correlated with a 5% increase in the likelihood of believing disinformation narratives (H4a). This 5% increase is replicated when we look at the frequency of Russian television consumption (H5). We find in an almost uncanny mirror image that viewing Ukrainian television and viewing it frequently both decrease the likelihood by 6% at a statistically significant level.

Figure 3.

Correlates of believing EU or the United States organized EuroMaidan (Hale et al., 2014).

Table 6.

Full Effects of Variables on Probability of Believing that the EU or the United States Are the Main Organizers of the EuroMaidan (Hale et al., 2014).

	Demographic variablesStage 1	Declared demographic variablesStage 2	Socio-economic variablesStage 3	EM protest participation variablesStage 4	Old mediaStage 4	New mediaStage 5	Robustness check(Stage 6)
All significant control variables
East	0.12*	0.09**	0.08*	0.12**	0.05	0.05	0.05
Education	0.06*	0.06*	0.06*	0.05*	0.06*	0.06*	0.06*
Nation forced choice Ukrainian (Ukrainian Nationality)		–0.04*	–0.03	–0.06*	–0.02	–0.02	–0.03
Transition loser			0.05**	0.06**	0.04**	0.04**	0.05**
Explanatory variables
Watches Ukrainian TV channels					–0.06**	–0.06**
Watches Russian TV channels					0.05**	0.05**
Does not watch TV					–0.03	–0.02	–0.01
Not user of Internet						0.01	0.00
Odnoklasnyky						0.02
VKontakte						–0.03
Facebook						0.03
Uses Internet for news						0.00
Ukrainian TV frequency							–0.06*
Russian TV frequency							0.05**
VKontakte frequency							–0.04
Facebook frequency							0.03
Odnoklasnyky frequency							0.04
Internet news frequency							–0.02
N	2,015	2,015	2,015	2,015	2,015	2,015	2,015

95% confidence intervals in brackets. Calculated using logit model.

p < .05; **p < .01

Non-significant variables omitted, see Table A2 in Supplemental Appendix.

Aside from television viewership habits, we also find that being a transition loser increases the likelihood of believing disinformation narratives. Somewhat surprisingly, higher education is also statistically significantly correlated with believing disinformation at a 6% increase. It is striking that no other control variables seem to be significant regarding disinformation narratives.

We note that the non-significant finding for VKontakte usage being associated with a greater likelihood in believing disinformation about the EuroMaidan confirms hypothesis H4c regarding Russian-based information sources. We suggest this is partly due to a misplaced assumption that VKontakte content is consumed in similar manner to Russian broadcast media. In fact, this suggests that VKontakte usage works like other, Western-based social media (a part of our H1) in which similar information bubbles and content silos can develop.

Thus, we are consistently able to confirm both our “old” and “new” media hypotheses (H1, H2). We can also confirm our Russian television hypothesis and frequency of viewership hypotheses (H4a) and reject competing hypotheses found in the literature about the role of Russian social media (H4b). Finally, we are only able to partially reject the “affective polarization” hypotheses (H6, H7, H8) as we find limited evidence to support that region of residence, civic identity, and being a loser of transition align with our dependent variables in the way expected above. Much of these effects are also mediated or absorbed by our main variables of interest and thus, we feel this mixed result only further highlights the importance of Russian television viewership on holding positive views of or believing in disinformation about the EuroMaidan.

Conclusion

We have provided substantial evidence that greater skepticism is necessary toward claims about the importance of social media during periods of mass mobilization. In the context of the EuroMaidan and the period immediately following mass mobilization, we find that social media usage was not noticeably associated with protest mobilization, general views on the protests, or belief in common disinformation narratives. Instead, we consistently find that “old” broadcast media consumption is significantly and substantively associated with these three aspects of public behavior and belief.

We confirm hypotheses that Russian television is strongly and negatively associated with protest participation, as well as in general negative attitudes toward the EuroMaidan and an increased likelihood of believing disinformation narratives regarding malign, foreign control over protest organization. We also find that consuming Ukrainian television makes it less likely that one buys into such disinformation. Our analysis also suggests that traditional macro-cleavages in Ukraine regarding identity, economic grievances, and region play strong roles in informing participation and views.

In one sense, the confirmation of our hypotheses regarding social media should be a relief to scholars. As research moves forward trying to determine the impact of “new” media and the massive technological and communicational change that it has engendered, it is helpful to know that the world has not yet dramatically transformed, even for events that the popular press dubs “hashtag” or “internet” revolutions. Our results suggest that some reservations are warranted over grand claims about the strong association of social media usage to particular political behaviors, even in extraordinary political contexts.

Given the correlational nature of this study, we cannot provide firm causal inferences for the associations we provide. We have strong theoretical reasons for believing that the direction of effect aligns with our results, but only future research will allow for us to fully assuage concerns about endogeneity. At the same time, we maintain that we have identified and confirmed an important element in the broader media environment—the continued consumption of broadcast television—that clearly is importantly related to both political action and beliefs during periods of crisis. We suggest that future research analyzing media effects on politics should ensure to take television and other “old” media into account.

We cannot adjudicate whether the lack of association between social media and protest behavior and views is a function of the specific Ukrainian context. Although Ukraine is a non-Western state located on the periphery of the European core, it is also a highly educated, post-industrial country with significant Internet penetration. We see no reason to believe that the theoretical insights from this study could not help researchers study protest in other medium-income countries elsewhere in Eastern Europe or in other turbulent regions such as the Middle East or Latin America. Social media may be ubiquitous, but instances of mass mobilization remain perhaps beyond its significant reach, for the time being.

Supplemental Material

sj-pdf-1-sms-10.1177_2056305121999656 – Supplemental material for Mobilization, Mass Perceptions, and (Dis)information: “New” and “Old” Media Consumption Patterns and Protest

Supplemental material, sj-pdf-1-sms-10.1177_2056305121999656 for Mobilization, Mass Perceptions, and (Dis)information: “New” and “Old” Media Consumption Patterns and Protest by Olga Onuch, Emma Mateo and Julian G. Waller in Social Media + Society

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research,authorship,and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research,authorship,and/or publication of this article: This work was supported by the National Science Foundation (RAPID Grant Scheme) under Grant SES-1445194.

ORCID iD

Emma Mateo

Supplemental Material

Supplemental material for this article is available online.

Author Biographies

Olga Onuch (DPhil,Oxford) is an associate professor in Politics [Senior Lecturer] at the University of Manchester,until 2020 she was an Associate Fellow in Politics at Nuffield College,at the University of Oxford and currently is an associate of HURI at Harvard. Her research centers on the comparative study of protest (elections,migration,and identity) in Eastern Europe and Latin America,and also in inter-regional comparative analysis. She is currently PI of the “MOBILSE” project exploring the micro-foundations of out-migration and mass protest,and the “Identity and Borders in Flux” project on identity,public opinion,and political behavior in Ukraine. She was co-PI of UCEPS the NSF funded project that this research is based on.

olga.onuch@manchester.ox.ac.uk

Twitter: @oonuch

Emma Mateo is a DPhil (PhD) Candidate in Sociology at the University of Oxford. Her research interests include geospatial protest diffusion,the media,slogans and symbolism of protest,and the relationship between protest and democratic development,with a particular focus on the post-Soviet space.

emma.mateo@balliol.ox.ac.uk

Twitter: @emm_mateo

Julian G. Waller is a PhD Candidate in Political Science at George Washington University and an Associate Research Analyst at the Center for Naval Analyses. His academic research focuses on authoritarian politics,historical and comparative legislative studies,and the dynamics of post-Soviet politics.

jgwaller@gwmail.gwu.edu

Twitter: @JulianWaller

References

Allcott

Gentzkow

(2017). Social media and fake news in the 2016 election. Journal of Economic Perspectives, 31(2), 211–236.

Arel

(1995). Language politics in independent Ukraine: Towards one or two state languages? Nationalities Papers, 23(3), 597–622.

Bakshy

Messing

Adamic

L. A.

(2015). Exposure to ideologically diverse news and opinion on Facebook. Science, 348(6239), 1130–1132.

Barberá

Jost

J. T.

Nagler

Tucker

J. A.

Bonneau

(2015). Tweeting from left to right: Is online political communication more than an echo chamber? Psychological Science, 26(10), 1531–1542.

Barrington

L. W.

Faranda

(2009). Reexamining region, ethnicity, and language in Ukraine. Post-Soviet Affairs, 25(3), 232–256.

Barrington

L. W.

Herron

E. S.

(2004). One Ukraine or many? Regionalism in Ukraine and its political consequences. Nationalities Papers, 32(1), 53–86.

BBC Trending. (2013, November 22). European square: Ukraine’s protest hashtag? BBC News. http://www.bbc.com/news/blogs-trending-25050141

BBG Gallup. (2014). Contemporary media use in Ukraine. Broadcasting Board of Governors Gallup. https://www.bbg.gov/wp-content/media/2014/06/Ukraine-research-brief.pdf

Beissinger

M. R.

Jamal

A. A.

Mazur

(2015). Explaining divergent revolutionary coalitions: Regime strategies and the structuring of participation in the Tunisian and Egyptian revolutions. Comparative Politics, 48(1), 1–24.

10.

Benkler

(2018). Network propaganda: Manipulation, disinformation, and radicalization in American politics. Oxford University Press.

11.

Bennett

W. L.

(2013). The logic of connective action: Digital media and the personalization of contentious politics. Cambridge University Press.

12.

Bezruk

(2014, March 3). Russian information warfare against the Maidan’s far-right aimed to discredit protests in Ukraine. Euromaidan Press. http://euromaidanpress.com/2014/03/03/russian-information-warfare-against-the-maidans-far-right-aimed-to-discredit-protests-in-ukraine/

13.

Boxell

Gentzkow

Shapiro

(2020). Cross-country trends in affective polarization (No. w26669). National Bureau of Economic Research. https://doi.org/10.3386/w26669

14.

Boyle

M. P.

Schmierbach

(2009). Media use and protest: The role of mainstream and alternative media use in predicting traditional and protest participation. Communication Quarterly, 57(1), 1–17.

15.

Campbell

(1980). The American voter (Unabridged ed.). University of Chicago Press.

16.

Chaffee

S. H.

Metzger

M. J.

(2001). The end of mass communication? Mass Communication and Society, 4(4), 365–379. https://doi.org/10.1207/S15327825MCS0404_3

17.

Chesney

Citron

(2019). Deepfakes and the new disinformation war: The coming age of post-truth geopolitics. Foreign Affairs, 98(1), 147–155.

18.

Clarke

Fredin

(1978). Newspapers, television and political reasoning. Public Opinion Quarterly, 42(2), 143–160.

19.

Colton

T. J.

(2000). Transitional citizens: Voters and what influences them in the new Russia. Harvard University Press.

20.

Crabtree

Darmofal

Kern

H. L.

(2015). A spatial analysis of the impact of West German television on protest mobilization during the East German revolution. Journal of Peace Research, 52(3), 269–284.

21.

Djerf-Pierre

Shehata

(2017). Still an agenda setter: Traditional news media and public opinion during the transition from low to high choice media environments. Journal of Communication, 67(5), 733–757.

22.

Dyczok

(2014). Information wars: Hegemony, counter-hegemony, propaganda, the use of force, and resistance. Russian Journal of Communication, 6(2), 173–176.

23.

Dyczok

Gaman-Golutvina

O. V.

(2009). Media, democracy and freedom: The post-communist experience. Peter Lang.

24.

EBU. (2017). Trust in Media 2017. European Broadcasting Union. https://www.ebu.ch/publications/trust-in-media-2017

25.

The Economist. (2017, May 19). Ukraine bans its top social networks because they are Russian. The Economist. https://www.economist.com/europe/2017/05/19/ukraine-bans-its-top-social-networks-because-they-are-russian

26.

Enikolopov

Petrova

Zhuravskaya

(2011). Media and political persuasion: Evidence from Russia. American Economic Review, 101(7), 3253–3285.

27.

EU vs DISINFORMATION. (2019a, March 9). Extreme nationalists staged coup and brought Poroshenko to power in Ukraine. https://euvsdisinfo.eu/report/extreme-nationalists-staged-coup-brought-poroshenko-to-power-in-ukraine/

28.

EU vs DISINFORMATION. (2019b, April 25). Euromaidan was a coup d’etat of the Right Sector and Ukrainian Nazis, supported by the CIA and Poland. https://euvsdisinfo.eu/report/euromaidan-was-a-coup-detat-conducted-by-the-right-sector-and-ukrainian-nazis-and-supported-by-the-cia-and-by-the-polish-government/

29.

Fedor

(2015). Introduction: Russian media and the war in Ukraine. Journal of Soviet and Post-soviet Politics and Society, 1(1), 1–12.

30.

Flaxman

Goel

Rao

J. M.

(2016). Filter bubbles, echo chambers, and online news consumption. Public Opinion Quarterly, 80(S1), 298–320.

31.

Flynn

D. J.

Nyhan

Reifler

(2017). The nature and origins of misperceptions: Understanding false and unsupported beliefs about politics. Political Psychology, 38(Suppl. 1), 127–150.

32.

Garrett

R. K.

(2006). Protest in an information society: A review of literature on social movements and new ICTs. Information, Communication & Society, 9(2), 202–224.

33.

Gentzkow

(2006). Television and voter turnout. The Quarterly Journal of Economics, 121(3), 931–972. https://doi.org/10.1162/qjec.121.3.931

34.

Giuliano

(2018). Who supported separatism in Donbas? Ethnicity and popular opinion at the start of the Ukraine crisis. Post-Soviet Affairs, 34(2–3), 158–178.

35.

Glaser

W. A.

(1965). Television and voting turnout. Public Opinion Quarterly, 29(1), 71–86.

36.

Gruzd

Tsyganova

(2015). Information wars and online activism during the 2013/2014 crisis in Ukraine: Examining the social structures of pro-and anti-Maidan groups. Policy & Internet, 7(2), 121–158.

37.

Hale

H. E.

(2013). Did the Internet break the political machine? Moldova’s 2009 “Twitter revolution that wasn’t.” Demokratizatsiya, 21(4), 481–506.

38.

Hale

H. E.

Colton

Onuch

Kravets

(2014). Ukrainian Crisis Election Panel Survey (UCEPS). https://ukraineelectionsurvey.com/

39.

Hale

H. E.

Shevel

Onuch

(2018). Believing facts in the fog of war: Identity, media and hot cognition in Ukraine’s 2014 Odesa tragedy. Geopolitics, 23(4), 851–881.

40.

Hayes

(2009). Has television personalized voting behavior? Political Behavior, 31(2), 231–260.

41.

Helmus

T. C.

Bodine-Baron

Radin

Magnuson

Mendelsohn

Marcellino

Bega

Winkelman

(2018). Russian social media influence: Understanding Russian propaganda in eastern Europe. Rand Corporation.

42.

Hilleary

(2014, March 14). Ukraine’s social media revolution years in the making. Voice of America. https://www.voanews.com/europe/ukraines-social-media-revolution-years-making

43.

Howard

P. N.

Bradshaw

(2018). The global organization of social media disinformation campaigns. Journal of International Affairs, 71(15). https://jia.sipa.columbia.edu/global-organization-social-media-disinformation-campaigns

44.

Iyengar

(1987). News that matters: Television and American opinion. University of Chicago Press.

45.

Iyengar

Lelkes

Levendusky

Malhotra

Westwood

S. J.

(2019). The origins and consequences of affective polarization in the United States. Annual Review of Political Science, 22, 129–146.

46.

Jaitner

(2015). Russian information warfare: Lessons from Ukraine. In Geers

(Ed.), Cyber warfare in perspective: Russian aggression against Ukraine (pp. 87–94). North Atlantic Treaty Organization, Cooperative Cyber, Defence Centre of Excellence.

47.

Jasper

J. M.

(1998). The emotions of protest: Affective and reactive emotions in and around social movements. Sociological Forum, 13, 397–424.

48.

Johnson

A. A.

LeFebvre

(2018). Contextual predictors of protest behavior on social media: A #Ferguson case study. Journal of Information Technology & Politics, 15(1), 50–65.

49.

Jones

Wayland

(2013). Activism or slacktivism? The role of social media in effecting social change [Research paper]. School of Engineering and Applied Science, University of Virginia.

50.

Jost

J. T.

Barberá

Bonneau

Langer

Metzger

Nagler

Sterling

Tucker

J. A.

(2018). How social media facilitates political protest: Information, motivation, and social networks. Political Psychology, 39(S1), 85–118.

51.

Khamis

Vaughn

Gould

P. B.

(2012, March 28). Beyond Egypt’s “Facebook revolution” and Syria’s “YouTube uprising”: Comparing political contexts, actors and communication strategies. Arab Media & Society. https://www.arabmediasociety.com/beyond-egypts-facebook-revolution-and-syrias-youtube-uprising-comparing-political-contexts-actors-and-communication-strategies/

52.

King

Schneer

White

(2017). How the news media activate public expression and influence national agendas. Science, 358(6364), 776–780.

53.

Kozachenko

(2014). How social media transformed pro-Russian nostalgia into violence in Ukraine. http://eprints.whiterose.ac.uk/114776/1/How_social_media_transformed_pro-Russian_nostalgia_into_violence_in_Ukraine.pdf

54.

Kulyk

(2001). The politics of ethnicity in post-Soviet Ukraine: Beyond Brubaker. Journal of Ukrainian Studies. https://ipiend.gov.ua/wp-content/uploads/2019/03/Kulyk-JUS-2001.pdf

55.

Kulyk

(2014). The media at the time of unrest: A report of a Maidan participant. Russian Journal of Communication, 6(2), 181–185.

56.

Kulyk

(2019). Identity in transformation: Russian-speakers in post-Soviet Ukraine. Europe-Asia Studies, 71(1), 156–178.

57.

Laitin

D. D.

(1998). Identity in formation: The Russian-speaking populations in the near abroad. Cornell University Press.

58.

Leeper

T. J.

Slothuus

(2014). Political parties, motivated reasoning, and public opinion formation. Political Psychology, 35, 129–156.

59.

Levendusky

M. S.

(2018). Americans, not partisans: Can priming American national identity reduce affective polarization? The Journal of Politics, 80(1), 59–70.

60.

Lewandowsky

Ecker

U. K. H.

Cook

(2017). Beyond misinformation: Understanding and coping with the “post-truth” era. Journal of Applied Research in Memory and Cognition, 6(4), 353–369.

61.

Lodge

Taber

C. S.

(2013). The rationalizing voter. Cambridge University Press.

62.

Lokot

(2013, December 6). As Ukraine’s protests escalate, #Euromaidan hashtag lost in a sea of information. Global Voices. https://globalvoices.org/2013/12/06/as-ukraines-protests-escalate-euromaidan-hashtag-lost-in-sea-of-information/

63.

Lotan

Graeff

Ananny

Gaffney

Pearce

(2011). The revolutions were tweeted: Information flows during the 2011 Tunisian and Egyptian revolutions. International Journal of Communication, 5, Article 31.

64.

MacKuen

Marcus

Neuman

W. R.

Miller

P. R.

(2010). Affective intelligence or personality? State vs. trait influences on citizens’ use of political information. Behavioral & Experimental Economics e-Journal. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1643468

65.

Mason

(2016). A cross-cutting calm: How social sorting drives affective polarization. Public Opinion Quarterly, 80(S1), 351–377.

66.

McCombs

(2014). Setting the agenda: Mass media and public opinion (2nd ed.). Polity Press.

67.

McCombs

Shaw

(1972). The agenda-setting function of mass media. The Public Opinion Quarterly, 36(2), 176–187.

68.

McLeod

D. M.

Detenber

B. H.

(1999). Framing effects of television news coverage of social protest. Journal of Communication, 49(3), 3–23.

69.

Mejias

U. A.

Vokuev

N. E.

(2017). Disinformation and the media: The case of Russia and Ukraine. Media, Culture & Society, 39(7), 1027–1042.

70.

Moroz

(2013, December 1). Vitaliy Moroz on Twitter: “Вчора найбільша активність в Твітері була твітів #євромайдан за годину”/Twitter4800. Twitter. https://twitter.com/insider_ua/status/407205906036686849

71.

Murphy

Link

M. W.

Childs

J. H.

Tesfaye

C. L.

Dean

Stern

Pasek

Cohen

Callegaro

Harwood

(2014). Social media in public opinion research: Executive summary of the aapor task force on emerging technologies in public opinion research. Public Opinion Quarterly, 78(4), 788–794.

72.

Nelson

J. L.

Webster

J. G.

(2017). The myth of partisan selective exposure: A portrait of the online political news audience. Social Media+ Society, 3(3), Article 729314.

73.

Newman

Fletcher

(2017). Bias, bullshit and lies: Audience perspectives on low trust in the media. Oxford: Reuters Institute for the Study of Journalism.

74.

Nisbet

E. C.

Kamenchuk

(2019). The psychology of state-sponsored disinformation campaigns and implications for public diplomacy. The Hague Journal of Diplomacy, 14(1–2), 65–82.

75.

Oates

(2016). Russian media in the digital age: Propaganda rewired. Russian Politics, 1(4), 398–417.

76.

Office of the United Nations High Commissioner for Human Rights. (2019). Report on the human rights situation in Ukraine 16 May to 15 August 2019. https://www.ohchr.org/Documents/Countries/UA/ReportUkraine16May-15Aug2019_EN.pdf

77.

Onuch

(2014). Who were the protesters? Journal of Democracy, 25(3), 44–51.

78.

Onuch

(2015a). “Facebook helped me do it”: Understanding the EuroMaidan protester “tool-kit.” Studies in Ethnicity and Nationalism, 15(1), 170–184.

79.

Onuch

(2015b). EuroMaidan protests in Ukraine: Social media versus social networks. Problems of Post-Communism, 62(4), 217–235.

80.

Onuch

Hale

H. E.

(2018). Capturing ethnicity: The case of Ukraine. Post-Soviet Affairs, 34(2–3), 84–106.

81.

Onuch

Sasse

(2016). The Maidan in movement: Diversity and the cycles of protest. Europe-Asia Studies, 68(4), 556–587.

82.

Onuch

(forthcoming). “Maidans” and movements: Legacies, innovations, and contention in independent Ukraine. In Stoltzfus

(Ed.) Rulers and Rebels: The Rise of Tyranny and the Power of the People. Bloomsbury.

83.

Papathanassopoulos

Coen

Curran

Aalberg

Rowe

Jones

Rojas

Tiffen

(2013). Online threat, but television is still dominant: A comparative study of 11 nations’ news consumption. Journalism Practice, 7(6), 690–704.

84.

Pariser

(2011). The filter bubble: How the new personalized web is changing what we read and how we think. Penguin Books.

85.

Peisakhin

Rozenas

(2018). Electoral effects of biased media: Russian television in Ukraine. American Journal of Political Science, 62(3), 535–550.

86.

Placek

M. A.

(2017). #Democracy: Social media use and democratic legitimacy in Central and Eastern Europe. Democratization, 24(4), 632–650.

87.

President of Ukraine. (2014, May 15). UKAZ PREZYDENTA UKRAYINY [ORDER OF THE PRESIDENT OF UKRAINE] (№. 133/20171332017). Ofitsiyne Internet-Predstavnytstvo Prezydenta Ukrayiny. https://www.president.gov.ua/documents/1332017-21850

88.

Prior

(2006). The incumbent in the living room: The rise of television and the incumbency advantage in U.S. house elections. Journal of Politics, 68(3), 657–673.

89.

Redlawsk

D. P.

(2002). Hot cognition or cool consideration? Testing the effects of motivated reasoning on political decision making. The Journal of Politics, 64(4), 1021–1044.

90.

Rogowski

J. C.

Sutherland

J. L.

(2016). How ideology fuels affective polarization. Political Behavior, 38(2), 485–508.

91.

Sørensen

R. J.

(2019). The impact of state television on voter turnout. British Journal of Political Science, 49(1), 257–278.

92.

Salzman

Aloisi

(2009). News media consumption and political participation in Central America: Causation and explanation. Journal of Spanish Language Media, 2, 46–75.

93.

Segerberg

Bennett

W. L.

(2011). Social media and the organization of collective action: Using Twitter to explore the ecologies of two climate change protests. The Communication Review, 14(3), 197–215.

94.

Shapiro

R. Y.

Jacobs

L. R.

(2011). The Oxford handbook of American public opinion and the media. Oxford University Press.

95.

Slothuus

De Vreese

C. H.

(2010). Political parties, motivated reasoning, and issue framing effects. The Journal of Politics, 72(3), 630–645.

96.

Snow

Benford

R. D.

(1988). Ideology, frame resonance, and participant mobilization. International Social Movement Research, 1, 197–217.

97.

Spaiser

Chadefaux

Donnay

Russmann

Helbing

(2017). Communication power struggles on social media: A case study of the 2011–12 Russian protests. Journal of Information Technology & Politics, 14(2), 132–153. https://doi.org/10.1080/19331681.2017.1308288

98.

Spohr

(2017). Fake news and ideological polarization: Filter bubbles and selective exposure on social media. Business Information Review, 34(3), 150–160.

99.

Stepanchuk

(2014, August 29). Poroshenko ta yoho “Dyvanna sotnya.” Ukrainska Pravda. https://www.pravda.com.ua/columns/2014/08/29/7036126/

100.

StopFake.org. (2014a, March 1). Stop fake EuroMaidan. https://www.stopfake.org/en/tag/euromaidan/

101.

StopFake.org. (2014b, March 5). Fake: USA gave Ukraine $ billion for coup d’etat5. https://www.stopfake.org/en/324/

102.

StopFake.org. (2014c, April 14). Russian fiction the sequel: 10 more false claims about Ukraine. https://www.stopfake.org/en/russian-fiction-the-sequel-10-more-false-claims-about-ukraine/

103.

Stroud

N. J.

(2008). Media use and political predispositions: Revisiting the concept of selective exposure. Political Behavior, 30(3), 341–366.

104.

Stroud

N. J.

(2010). Polarization and partisan selective exposure. Journal of Communication, 60(3), 556–576.

105.

Szostek

(2014a). Russia and the news media in Ukraine: A case of “soft power”? East European Politics & Societies and Cultures, 28(3), 463–486. https://doi.org/10.1177/0888325414537297

106.

Szostek

(2014b). The media battles of Ukraine’s EuroMaidan. Digital Icons, 11, 1–19.

107.

Szostek

(2018). Nothing is true? The credibility of news and conflicting narratives during “information war” in Ukraine. The International Journal of Press/Politics, 23(1), 116–135.

108.

Taber

C. S.

Lodge

(2006). Motivated skepticism in the evaluation of political beliefs. American Journal of Political Science, 50(3), 755–769.

109.

Tolbert

C. J.

McNeal

R. S.

(2003). Unraveling the effects of the Internet on political participation? Political Research Quarterly, 56(2), 175–185.

110.

Trottier

Fuchs

(2014). Social media, politics and the state: Protests, revolutions, riots, crime and policing in the age of Facebook, Twitter and YouTube. Routledge.

111.

Tucker

J. A.

(2006). Regional economic voting: Russia, Poland, Hungary, Slovakia, and the Czech Republic, 1990-1999. Cambridge University Press.

112.

Tufekci

(2017). Twitter and tear gas: The power and fragility of networked protest. Yale University Press.

113.

Tufekci

Wilson

(2012). Social media and the decision to participate in political protest: Observations from Tahrir square. Journal of Communication, 62(2), 363–379.

114.

Vitak

Zube

Smock

Carr

C. T.

Ellison

Lampe

(2010). It’s complicated: Facebook users’ political participation in the 2008 election. Cyberpsychology, Behavior, and Social Networking, 14(3), 107–114. https://doi.org/10.1089/cyber.2009.0226

115.

Vlasov

Leonard

(2014). Ukraine’s Facebook revolution: 1 year later. AP News. https://apnews.com/article/8e65bcdd8b694ac8ab5fd82f5b90f7b6

116.

Walker

Orttung

(2014). Breaking the news: The role of state-run media. Journal of Democracy, 25(1), 71–85.

117.

Wardle

(2017, February 16). Fake news. It’s complicated. First Draft. https://firstdraftnews.org:/latest/fake-news-complicated/443

118.

Wei

Hindman

D. B.

(2011). Does the digital divide matter more? Comparing the effects of new media and old media use on the education-based knowledge gap. Mass Communication and Society, 14(2), 216–235.

119.

Wellman

Haase

Witte

Hampton

(2001). Does the Internet increase, decrease, or supplement social capital? Social networks, participation, and community commitment. American Behavioral Scientist, 45(3), 436–455.

120.

Westen

Blagov

P. S.

Harenski

Kilts

Hamann

(2006). Neural bases of motivated reasoning: An fMRI study of emotional constraints on partisan political judgment in the 2004 US presidential election. Journal of Cognitive Neuroscience, 18(11), 1947–1958.

121.

Wojcieszak

Garrett

R. K.

(2018). Social identity, selective exposure, and affective polarization: How priming national identity shapes attitudes toward immigrants via news selection. Human Communication Research, 44(3), 247–273.

122.

Yandex. (2014). Oglyad sotsіalʼnikh myeryezh і Tvіttyera v Ukraїnі. [Overview of social networks and Twitter in Ukraine] http://download.yandex.ru/company/Yandex_on_UkrainianSMM_Summer_2014.pdf

123.

Youmans

W. L.

York

J. C.

(2012). Social media and the activist toolkit: User agreements, corporate interests, and the information infrastructure of modern social movements. Journal of Communication, 62(2), 315–329.

124.

Zhukov

Y. M.

(2016). Trading hard hats for combat helmets: The economics of rebellion in eastern Ukraine. Journal of Comparative Economics, 44(1), 1–15.

125.

Zhukov

Y. M.

(2017, April). Warfare in a post-truth world: Lessons from Ukraine. Ponarseuarasia: Policy Memos. http://www.ponarseurasia.org/memo/warfare-post-truth-world-lessons-ukraine

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.40 MB

Mobilization,Mass Perceptions,and (Dis)information: “New” and “Old” Media Consumption Patterns and Protest

Abstract

Keywords

Introduction

EuroMaidan: Overview

Media, Mobilization, and Mass Perceptions: Framing the Analysis

Media Consumption and Political Participation

Media Consumption and Protest Participation

Media Consumption and Perceptions of Protest

Media Consumption and Buying into Disinformation

Data and Operationalization

Dependent Variables

Independent Variables

Old Media

New Media

Frequency of TV Watching

Control Variables

Causal Sequencing and Analytical Approach

Results and Discussion

Protest Participation

Positive Views of the EuroMaidan

Belief in Disinformation about the EuroMaidan

Conclusion

Supplemental Material

sj-pdf-1-sms-10.1177_2056305121999656 – Supplemental material for Mobilization, Mass Perceptions, and (Dis)information: “New” and “Old” Media Consumption Patterns and Protest

Footnotes

Declaration of Conflicting Interests

Funding

ORCID iD

Supplemental Material

Author Biographies

References

Supplementary Material