Information Sources Shared on Facebook and Networking by Populist Leaders and Populist Parties in Slovakia

The case study undergoes an analysis of the FaceBook (FB) activity of two Slovak populist entities, one lead by Boris Kollár and OĽaNO movement. The aim was to explore whether and how they construct a populist network and draw on media sources that are out of the mainstream or associated with a populist style. Both selected entities were highly active and quite successful in their communication on FB throughout most of the last five years at least. The study finds that these entities publish sources and connect with other FB pages that mostly confirm their parties’ positions and alliances. Yet there was no preference for alternative sources, understood as disseminating hyper-partisan or fake news and hoaxes, often associated with populist parties. Rather, OĽaNO had some reciprocity in media visibility with mainstream liberal media, in the sense that the party drew on liberal mainstream sources, which also covered the parties’ activities. In sharing posts, Boris Kollár had the most intensive dissemination network, while movements WAF and OĽaNO were very similar in terms of size of their network. There was a rather low cross-promotion. OĽaNO supporters were more appreciative of the efforts made by murdered journalist Ján Kuciak, and showed a stronger focus against Smer-SD (Direction-Social Democracy). In contrast, Kollár was associated more with pro-Christian, national and social rhetoric.


Introduction 1
This study is focused on information sources shared on FaceBook (FB) and networking on this social platform by selected populist leaders and populist parties. The study has two analytical parts. In the first part, it provides a classification of sources shared or ignored by populist 1 The data was coded by two Slovak coders -Viera Žúborová and Lubica Adamcová and then checked and an agreement was reached between the two coders. This project has received funding from the European Union's Horizon 2020 research and innovation programme under grant agreement No 822590 (DEMOS). Any dissemination of results here presented reflects only the consortium's (or, if applicable, author's) view. The Agency is not responsible for any use that may be made of the information it contains. Štúdie a analýzy respectively, since March 2020). Thus, the key research questions are even more relevant: do they "like" alternative or any other (mainstream) media on their FB pages and what are their networks on FB?
Moreover, the selection of these two parties for further analysis was also relevant from the perspective of social media use in political communication. OĽaNO had the highest number (3,201,755) of all types of interactions on FB among political parties throughout 2020. WAF was third in the number of interactions (1,102,550), the second place was occupied by Smer-SD (1,232,700). Even more impressive was the total number of interactions of Igor Matovič in 2020 -6,370,753 -more than all other members of the Cabinet combined (Tóth 2021). Similarly, the highest engagement of the followers during 2019 campaign to the European Parliament was recorded on the FB page of OĽaNO (52,000), followed by WAF (29,000). Similarly, the highest number of posts on FB was presented by OĽaNO (129), followed by WAF (88) (Oravcová, Plenta and Vicenova 2019). Both parties were among the first (OĽaNO) or third (WAF) most popular parties on FB based on interactions during the six weeks of campaign before 2020 general elections (Klingová et al., 2020). Smer-SD was the second most popular party on FB. However, OĽaNO was the clear leader, having about the same popularity on FB as the three following political parties combined. WAF published 22 times more contributions on FB than on its website during the official 2019 election campaign before elections to the EP (Struhár, 2019).
For further analysis, we followed a common theoretical-methodological approach described in a separate chapter. Before going into the analysis, we deemed it necessary to have an overview of previous local research on populism and political communication in the country.

Research Review
There is no specific research on this topic but there are quite many studies that tackle aspects of parties' political communication on social media. We provide an overview of the most relevant studies. Major political parties started to join FB around 2009 (Bardovič 2018). The general use of social media by political parties could be noticed for the first time before 2010 general elections when all parties used at least some social media. In general, the first time social media played an important electoral role in Slovakia was in the 2016 parliamentary elections according to sociologist Slosiarik 5 . Before that, during the 2012 -2014 period, the online communication strategy of political parties was rather one sided. Political parties mostly just released some information that was further disseminated on communication channels. Among the most interactive on FB were Freedom and Solidarity (Slovak: Sloboda a Solidarita, SaS) and OĽaNO.
There were two dominant tools of communication and information dissemination -party websites and FB, with secondary roles of YouTube and Twitter (Zagrapan 2016). The FB pages of parties were used mostly to reach the desired audience (Zagrapan 2019). In retrospect, the most successful trend in FB activities could be noticed in case of OĽaNO, while WAF was atypical with rather rapid initial increase of its followers. WAF used the FB following of its founder, 5 Sme, December 31, 2016, p.2. Boris Kollár, before the 2016 general elections. Communication activities of the Kotleba - People's Party-Our Slovakia (ĽSNS) were the most often analysed by local scholars. ĽSNS was well represented on FB, with many FB pages for local branches, allowing the quick promotion of any party activity on a local and micro-local level (Zagrapan 2019). The ĽSNS party utilized the politics of "Othering" through dissemination of populist and nativist frames, while using FB for self-organization between 2013 and 2015. Its discourse was spread through networks of active fans (Kluknavská & Hruška 2019). ĽSNS used the FB profile primarily for the presentation and promotion of Marián Kotleba, the chairman of the party, as a candidate for president in the upcoming 2018 elections. Identically, Bardovič (2018) found that the communication of ĽSNS was directed significantly onto the leader and is realized through the regional, district or fan pages of ĽSNS. Only a few MPs of ĽSNS have individual profiles or pages on social network sites. It should be mentioned that FB blocked key pages of ĽSNS already in 2017. 6 This was the only party that faced these communication challenges repeatedly.
Content-wise, Smer-SD, Most-Híd and SaS used FB profiles more intensively for reflection of their sectoral interests before the 2016 general elections. Communication of the SaS political party was characterised by a focus on criticising the government coalition, particularly Smer-SD and the SNS. Most-Híd periodically published posts about its ministers´ activities, while Smer-SD was mainly trying to communicate its key political topics (Garaj 2018). SNS before general elections in 2016 communicated on FB less frequently than ideologically similar parties in Slovakia and Czech Republic under the same condition. ĽSNS linked information from ideologically close webs, while SNS prioritised own electoral theses (Filipec, Garaj, Mihálik 2018).
If we focus on popularity trends, in August 2016, SaS had the highest number of FB fans, followed by ĽSNS and newly established WAF -using Fanpage Karma (Rózsa 2016). In 2017, SaS, OĽANO and WAF were the most popular parties on FB, with growing online popularity of the latter two parties (Tím Digitálka 2017). In 2018, SaS, OĽaNO and WAF had the highest number of "likes" on FB while ĽSNS, SaS and OĽaNO had the highest number of subscribers on YouTube. For Instagram, the highest number of followers was held by OĽaNO, WAF and SaS, while Twitter seemed to be relevant only for SaS (Bardovič 2018). In general, then strongest coalition party, Smer-SD, had, together with its coalition partner at the time, Most-Híd, a rather low fan base on FB, while SaS with a relevant fan base did not seem to utilise this base sufficiently (Tančinová 2018).
If we consider impact, then the 2017 regional elections showed that activity on FB possibly translated to election results only in case of OĽaNO and to a lesser degree for SaS. Neither WAF nor ĽSNS profited much electorally from their activities on FB. However, this was much impacted by the type of electoral system and for example lower interest of WAF leader in these elections (Tím Digitálka 2017). Ďurman (2019) has shown that the most expensive campaign on FB was carried out by Smer-SD in 2019, while ĽSNS did not invest any money in FB campaigns. Other major parties could be in most cases located in-between these two extremes. Yet ĽSNS was relatively successful and Smer-SD fared much worse than four years earlier (of course, there were other factors at play here, too). Murray Svidroňová, Kaščáková and Bambuseková (2019) analysed the FB pages and profiles of all 15 candidates for the post of President of the Slovak Republic in 2018. Using the Facepager tool they calculated the number and type of posts and interactions (i.e. number of likes and shares) for the candidates by calculating interaction rate. It was suggested that there is a big potential of social media use in political marketing.
Zuzana Čaputová, the winner of the elections had the highest number of "likes" and "comments", but only the second highest number of shares. The overall FB interactions rate of the final winner in elections was only at 6-7 th shared place (this means that they reached equal value). Overall, the more individual candidates added posts to their FB pages or profiles, the more their followers who engaged in the discussions became aware of them. This increased their viewership and the number of comments received on their posts as well as the post sharing itself.
Arguably, among the most successful messages during the 2020 general elections campaign disseminated via social media and which were then taken over by legacy and alternative/controversial media were two videos produced by OĽaNO. 7 The first video had 1.6 million views, over 20,000 shares and about 32,000 reactions. The second video had 827,000 views, 20,000 likes and over 7,000 shares. 8 Finally, it should be mentioned that a group of researchers under the name of Katedra komunikácie monitors performance (interactions) of politicians, political parties and state authorities profiles on FB since 2020. 9

Analytical Part 1: Sources Shared by Populist Leaders
In this part we focus on the classification of sources shared or ignored by populist leaders in Slovakia in two selected periods in 2020. This part includes the examination of the dominant political/ideological orientation of the media sources shared. We then compare different types of coverage by type and period: electoral versus non-electoral and event-driven versus regular coverage. Finally, we ask what role the public media played in each of these periods, compared to commercial /private sources.
Information Sources Shared on Facebook and Networking by Populist Leaders... all posts sharing a link were digital sources) as well as for Boris Kollár (89% of all posts sharing a link were digital sources).
The hierarchy of media channels was the same for both: newspapers and magazines were the second most often shared source but with big distance from digital sources (with 8% by Kollár and 9% by OĽaNO) followed by TV channels (2% in Kollár's posts and 1% in OĽaNO's) and with almost total absence of radio: less than 0.5% for each.
Overall, it would seem that the diversity in terms of type of media channel, or better said lack of diversity, is surprisingly similar for both political opponents (Fig. 1).

Figure 1: Diversity of channels
Another common tendency was the clear preference for FB content over other types of digital information sources. Both Boris Kollár and OĽaNO prioritized FB sources over any other, irrespective of the period of coverage, with a similar frequency varying from over half of all posts sharing a link to over 80%, depending on the period analysed.
Overall, both Kollár and OĽaNO shared FB pages in 73% of all their posts with a link. However, the diversity of sources was clearly higher in Boris Kollár's case. Despite posting a lower number of links, it was almost half compared to OĽaNO. Among all 438 such posts, besides other FB pages, Kollár shared 39 unique sources. This was almost identical if compared to OĽaNO -with 33 unique sources in addition to FB, out of 754 posts.
Many of the most frequently shared sources overlapped: liberal-centre newspaper and portal sme.sk was the second most often shared (after FB sources) by Kollár -5% of all posts, and the 5 th most shared by OĽaNO -with 3% posts.
Then there was liberal-centre newspaper and portal dennikn.sk which was the 3 rd most often shared media source by Kollár -2% posts and the 3 rd by OĽaNO -5%, respectively. This sample was followed by online only centrist news portal with investigative team aktuality.sk (foreign owned) that was the 4 th most shared by OĽaNO -4% posts, and the 8 th by Kollár -in 1% of the posts. It had actually the same level of sharing as for conservative Christian news and current affairs postoj.sk and youtube.com platform (1% each of them for each of the two political actors).
However, there were sources that only one of the two populist entitiesm shared during the 13 months under study. The "alternative" news site hlavnydennik.sk was shared only by Boris Kollár, in 2% of his posts, another alternative news site hlavnespravy.sk was also shared minimally -just in 1% posts.  As for OĽaNO, the 2 nd most shared source was the party website -obycajniludia.sk (7% posts), which Kollár did not share (being a political adversary). There were two other examples of websites that only OĽaNO shared, in less than 1% posts: a little known online only commentary weekly tyzdennikkoment.sk and centre-right business weekly website etrend.sk.
A more in-depth analysis of the type of digital sources shared by each analysed subject shows similarities and differences that occurred (Fig. 4). There was a clear, shared, preference for content from the political party or members of the political party of each political actor analysed. This was more pronounced for OĽaNO -86% of all digital sources, and less so for Boris Kollár -67%.
The second most favourite source of digital information for both were news websites and blogs (8% Kollár, 7% OĽaNO).
Public authorities' websites or expert sources were not very popular, not even during the CO-VID-19 crisis. They were almost entirely missing on Kollár's FB, and on OĽaNO FB they made up only 1% of all posts. Nonetheless, it seems that Kollár used a higher variety of digital sources compared with OĽaNO.
It is also important to note that in 10% of the cases (Kollár) and almost in 20% of cases (OĽaNO) information sources did not fit the categories in the codebook, being difficult to code in a reliable way. Overall, the two populist actors have used their FB pages in all three analysed periods more as a self-promotion tool, to make themselves, their party and colleagues visible rather than to inform people on different issues.
The populist actors´ appeal to news websites was very low. In Kollár's case it was even lower than forin other non-news media. It was strikingly common for both that it was less frequent than expected the presence of independent or investigative journalism, non-profit or crowdfundingbased journalistic initiatives.
In the case of Boris Kollár, he has explained his attitude towards alternative news sources as follows: "If these portals such as "Hlavné správy" will bring truthful information, I have no problem sharing themit. However, I definitely won't share hate and fake news" (in Kern, 2020).
In terms of newspaper type (Fig. 5), as mentioned, there seemed to be a slight difference in preferences. Boris Kollár favoured quality or at least non-tabloid newspapers and magazines (74% posts), while OĽaNO shared tabloid content more frequently (63%), though the difference from non-tabloids was not that big considering shares were overall low.

Media registration
In terms of Assessing media registration, this proved to have been a difficult research task. Only about a quarter of the links were coded as registered media for each of the two political actors. Only 4% (Kollár), respectively 3% (OĽaNO) of the posts were identified as not registered officially as news media. However, most sources (around 70%) either could not be classified as such (for example because many were FB pages of different kinds -like politicians' pages) or because there was no easily available, transparent and free of cost access to data in this regard.
Information Sources Shared on Facebook and Networking by Populist Leaders...

The dominant political/ideological orientation of the media sources shared
A similarity between the two political actors analysed was that they seemed to share information sources that lean more towards the right side of the political spectrum (Fig. 6). However, there were also important differences. Boris Kollár seemed to show a much more ambiguous political and ideological leaning, not captured by the standard left-right spectrum. Those sources coded largely as "other" (69% posts) were mostly right-wing populist and anti-immigration, conservative, but some sources were also considered as disseminating disinformation or, as they call themselves, "alternative" websites (ex. hlavnespravy.sk, hlavnydennik.sk, napalete.sk, parlamentnelisty.sk, denniks.sk, napalete.sk). These widely seen (there is a certain liberal bias in Slovakia here) disinformation sources made up to 4% of the sources shared by Boris Kollár, and were usually shared regularly, but also appeared once during EP elections (hlavne.sk) and during the COVID-19 crisis (hlavnespravy.sk).
Kollár also showed a preference for the black humour / satiric Zomri page, which OĽaNO also shared, but only once.
OĽaNO leaned more towards the conservatives, rather than populist or anti-immigration sources. Its sources wouldere more traditionally be defined as center-right leaning defined. Its sources would more traditionally by defined as center-right leaning.
The only source considered center-left that was shared was newspaper Pravda, distributed by (Kollár).
Considering the (rather disputed) political / ideological orientation of Kollár and OĽaNO, the sources they shared seemed to by and large mirror it. There was not much preference for a pluralistic media resources landscape. In addition, only 19% of the sources shared were considered fully transparent for both actors. There was also the issue of the difficulty to assess transparency of ownership in the absence of data (especially in the case of the sources shared by Boris Kollár).

Electoral vs. non-electoral coverage, event vs. regular period
In terms of frequency of posting in different time periods, both political actors shared information sources more often during regular political time periods (an average of 37 sources shared per month by Kollár and 60 by OĽaNO). A difference between them was that OĽaNO kept a similar frequency during the COVID-19 crisis (59 sources shared per month), while Boris Kollár shared significantly less sources (29 per month), identical to the EP electoral campaign (30 sources/ month). This could reflect his new role of Speaker of the Parliament during this period. OĽaNO shared other information sources the least often during the EP campaign (51/month).
What stands out is that both Kollár and OĽaNO's over-relied on FB in all three periods. Boris Kollár shared information from other FB pages, accounts or groups especially during the start of COVID-19 crisis (in 82% of all posts), in 72% during the regular coverage and slightly less during the EP campaign (67% posts).
OĽaNO followed the same pattern, slightly more pronounced, sharing other FB sources in 89% of all COVID-related posts, 74% during regular coverage and 60% during the election campaign.
Analysis of FB pages prioritised by each of them during the COVID-19 crisis, Boris Kollár shared his party's page (WAF) in 38 out of 57 posts, while all other FB sources were shared three times (the case of Kollár's personal public FB account 10 ) or just once: Igor Matovič's page, a car rental company's page 11 , the page of Police of the Slovak Republic -Banská Bystrica Region 12 , regional news from Dlhé nad Cirochou a okolie, his political fellow Milan Krajniak's page 13 and Regionportal.
Other sources in addition to FB that Kollár shared once or twice during the COVID-19 months were also mostly of digital type, like: public wire agency portal teraz.sk, news and current affairs online only portal aktuality.sk, online news portal of private TV Markíza tvnoviny.sk, controversial -alternative news portal hlavnespravy.sk, pomahameludom.sk ("we are helping the people"), portal of private wire agency webnoviny.sk, one TV channel's website -prezenu.joj. sk and two newspaper websites business newspaper slovensko.hnonline.sk, and video section of liberal centre right newspaper Sme -video.sme.sk.
In other words, it was a mixture of quality and "alternative" as well as other types of media sources. This implies that, overall, Boris Kollár clearly prioritized digital sources (especially social media) over mainstream news. This is natural, considering how FB and other social media function.
This pattern is even more clear in the case of OĽaNO, that prioritized digital only sources in 105 out of all 117 posts that shared a media source during COVID crisis, and 101 out of these were from FB. However, the diversity of these pages was bigger for OĽaNO than Kollár. Similar to Kollár, most posts shared were from the party leader -Igor Matovič (24 out of 101), followed by other more known party members or candidates for being MPs: Jaro Naď -12 posts, Kristián Čekovský -6 posts, Eduard Heger -7 posts, Gábor Grendel -4 posts, Michal Šipoš -4 posts, etc.
OĽaNO shared its party website once during the COVID-19 crisis. There seemed to be slightly more diversity in the sources shared by OĽaNO compared to Boris Kollár, which may also be due to the difference in the types of pages: one is a party page and one is a more personal (but publicly available), individualized party leader page.
A similar tendency was kept during regular coverage, with somewhat more diversity due to the longer time interval and different topics approached.
On Kollár's FB page, 211 out of 292 posts that shared a media source originated from public pages: from his party WAF (138 posts), content made by himself (like videos) or from his other, personal account (34 posts), other pages: a native Slovak living in Italy and supporting also Matteo Salvini -Luboš Hrica (3 posts), Martin Petriska (3 times), Patrick Linhart (2 times), Petra Krištúfková (M.P) (3 times), Matteo Salvini's page (Italian right wing populist politician, shared once) and other party colleagues or supporters. Among his other preferred FB pages seemed to be the black humour zomriofficial page, regional state police pages (KRPZBB -also shared during the COVID period, and policiaslovakia).
Among newspapers, the most often shared were quality liberal (centre-right) media, and occasionally tabloids, conservative weekly and business newspaper: dennikn.sk (8 times), sme.sk Other digital sources commonly referred to were: quality online only news portal aktuality. sk (19 posts), general affairs registered online weekly tyzdennikkoment.sk (4 posts), the blog section of the conservative Christian website blog.postoj.sk (2 posts), portal on European affairs euractiv.sk (1 post), public wire agency portal teraz.sk (1 post), tabloid portal topky.sk (2 posts), alternative news portal europskenoviny.sk (2 posts), glob.zoznam.sk (2 posts), commercial TV portal tvnoviny.sk (1 post), and the party website obycajniludia.sk was shared in 26 posts. The black humour zomriofficial shared by Kollár was also shared by OĽaNO, but only once.
For OĽaNO, digital sources were again clearly prioritized, especially social media, but newspapers were also a somewhat common reference source, though 10 times less often than digital sources. The newspapers shared by OĽaNO were: quality liberal newspaper (both print and online version) dennikn.sk ( The only video channel (formally called TV) shared was TV.teraz.sk (1 post) on a public wire agency website.
Radio was the least favourite source of shared information: Kollár drew info from only one radio: traffic RadioExpres (twice, via FB and YouTube), during the regular period. The same radio -RadioExpres was shared once by OĽaNO during the standard period.
We could see so far that there were some overlaps and similar tendencies during the regular reporting period and special reporting period during COVID-19 crisis.
For the election period, 60 out of 89 posts sharing a media source on Boris Kollár's page made reference to another FB page/account, just like in previous cases. In most cases, Kollár shared his party's page (24 posts), uploaded content from his own page or personal account (18 posts Other digital sources included the same varied mixture of quality sources, tabloid sources and alternative sources: aktuality.sk (3 posts), teraz.sk (2 posts) and others appearing only once: Máme rádi Karla Gotta -fanclub of famous Czech pop singer, Mukli SK 14 (about persons in jail, but heavily covering politics), black humour page zomriofficial, blog on used cars shopping blog.autobazar.eu.
As for newspapers, only quality, mostly center-right, liberal, such as sme.sk (5 posts), hnonline.sk (4 post), dennikn.sk (2 posts) and centre-left spravy.pravda.sk (1 post) were shared during the electoral campaign. It appears that Boris Kollár limited the diversity of sources shared the most during the EP campaign, where he prioritized party-related sources, European allies from the radical right (Salvini, IDP, Movement for a Europe of Nations and Freedom -MENF), main-stream newspapers and two digital news websites which were common information sources for all periods and for both Kollár and OľaNO.
Kollár's communication during the EP campaign seemed to be somewhat more formal, though his preference for the dark humour website Zomri remains constant and visible even during these times.
As for OĽaNO, the diversity of the sources seemed to be even more restricted during the campaign. For example, 93 out of all 154 links shared were from other Facebook pages: 14 from different FB events, Miroslav Sopko, MP for OĽaNO (11 posts), Jaro Naď, MP for OĽaNO (9 posts Other digital sources shared by OĽaNO during the EP election campaign included a mixture of online only media: aktuality.sk (10 posts), tyzdennikkoment.sk (2 posts) and shared only once: blog.etrend.sk, blog.postoj.sk, jaronad.sk, navody.digital, redflags.slovensko.digital and the party website was again shared 23 times.
Only one TV website was shared by OĽaNO during the electoral campaign: bbc.com (1 post), and four newspapers: dennikn.sk (9 posts), sme.sk (10 post), cas.sk (1 post), independent.co.uk (1 post). Similar to Kollár, OĽaNO also seemed to have formalized its communication and source-sharing during the EP elections.
Overall, there were common tendencies in the type of sources prioritized during each interval by both Kollár and OĽaNO, with a clear predominance of digital sources (including social media), that can be found in 88% to 95% of all posts during each period. This can be explained by the fact that the media platform used for communication is itself digital (FB pages of the two political actors), but it also seems to support the literature that emphasises that social media and digital media are indeed a vehicle for promoting populists.
Interestingly, Kollár draws on digital sources especially during the COVID-19 crisis (95%), while during both regular and electoral times, digital sources were slightly less frequent (88%). However, these differences can be seen as somewhat negligible.
For OĽaNO, the differences were smaller: digital sources were shared in 92% of the electoral campaign sample, 90% in COVID-related interval and 89% during regular coverage. Similarly, OĽaNO draws on newspapers most equally often irrespective of the period: 9% in non-events times, 8% during the COVID crisis and 7% during elections.
On the other hand, the discrepancies for Boris Kollár were more clear but still relatively marginal: newspapers were shared mostly during elections (12% posts), less often regularly (8%) and, somewhat surprisingly, the least often during the pandemic (2%). This last low number can be perhaps explained by the fact that Kollár became too busy in his post of Speaker of the Parliament.
A very clear tendency that emerged was the big absence of radio as source of information for both political actors. Only one radio channel was shared by both -traffic radio RadioExpres, two times by Kollár and once by OĽaNO, and all were within regular reporting period. TV channels were also very infrequently used during all analysed periods. When they did represent a source, it was mostly during the COVID-19 pandemic: in 4% of Kollár's posts and 3% of OĽaNO's. These findings are most probably also due to the nature of the communication environment: digital communication draws on other digital sources, due to proximity. Further studies could delve deeper by checking whether these political actors' communication via TV or radio is similar or different to their online presence. If there is substantial difference in what and how news is reported by digital versus traditional media, this could potentially generate two separate public spheres that have very different views on the surrounding political realities. Another difference that would be expected, drawing on Europeanization literature (Koopmans, Erbe & Meyer, 2010;Koopmans, 2010), is that events with a European or international dimension should draw more Europeanized or globalized coverage. To test this, we checked what types of media sources were more often cited during the three intervals. The results seem to confirm this hypothesis only to some extent (Fig. 9). During regular coverage it was the least common that European or international sources were shared and national sources clearly predominated. This was valid for Boris Kollár (4% of all posts) and to some extent, for OĽaNO (13% of posts). Interestingly, EP elections did not generate more Europeanized coverage than during regular coverage, neither for Kollár, nor for OĽaNO. But where the tendency becomes obvious it was with the pandemic-related information, which draws the most international coverage for both political actors (7% Kollár, 25% OĽaNO). Remarkably almost absent during all periods were regional and local sources, which only became somewhat more frequent in Kollár's posts during the COVID-19 crisis (4%). Among the European or international sources shared by the two political actors were Matteo Salvini, Czech media sources, BBC, Bloomberg, Euractiv, Independent or Greenpeace.
What role did the public service media play in each of these periods, compared to commercial /private sources?
In absolute numbers, the role of public service media was very low, almost negligible, in the communication of both populist actors, who each shared links to public information sources in around 1% of their posts. Common references were made only to the (PSM) Radio and Television of Slovakia (spravy.rtvs), cited by OĽaNO during the COVID-19 crisis and by Kollár during regular coverage. Kollár also shared a link to the ČT24 (PSM Czech language news TV channel) and to the FB page of the Slovak police (policiaslovakia), all during regular coverage.
OĽaNO, on the other hand, showed a clear tendency of drawing on publicly owned sources of information during the COVID-19 crisis (8 out of all 10 such references were during this time). They shared the FB pages or website of the Government Office, Parliament and other public institutions (Ministry of the Interior under political influence of OĹaNO, National Council of the Slovak Republic/ Parliament, Trnava self-governing region under political influence of OĽaNO). During the regular and elections periods, OĽaNO shared the page of Žilina Selfgoverning Region (where it had its political representant). These differences may be explained only by different positions of both actors -while OĽaNO was more prominently represented in the government, Kollár became Speaker of the Parliament.

Analytical Part 2: Network analysis of sources that shared populist leaders' posts
For network analysis, we focused on reciprocity of the network, its centrality and the main promoters of messages. In this part we conducted the network analysis adding also the public page of Sme Rodina (We Are Family -WAF), Boris Kollár's party. We were interested to explore the overlap between the two and the sources that were uniquely connected to one or the other.
Based on the CrowdTangle data (Mancuso et al, 2020;Marincea, 2020) analysed with No-deXL 15 , we constructed a directed graph with 1022 vertices (nodes) and 1035 unique edges (unique connections between nodes) out of a total number of 11,161 edges 16 . In the center were three main vertices: the public pages of WAF, Boris Kollár and OĽaNO (Fig. 10). The network represents all public pages that have shared posts from the three pages, between January 2019-April 2020 and the red arrows show reciprocity in connections.
The general overview (Fig. 10) shows that among the three pages, Boris Kollár had the most intensive dissemination network 17 , while WAF and OĽaNO were very similar in terms of size of their network. This seems to confirm the literature on populist communication which argues that personalization through charismatic leaders is often a successful strategy in gaining visibility. 15 Version 1.0.1.418. 16 The total number of edges represents each time one public page shared a post from one of the 3 pages under study. This also includes the pages shared by the 3 pages, in order to identify reciprocity.Additional graph metrics: graph density -0.0016, Average Geodesic Distance -2.34, Maximum Geodesic Distance (diameter) -3, Reciprocated Vertex Pair Ratio -0.034 and Reciprocated Edge Ratio -0.065. 17 Kollár was shared 4721 times by 639 different public pages, while his party Sme Rodina -3399 times (by 472 unique Facebook pages) and OĽaNO -2985 (by 472 unique FB pages).
Information Sources Shared on Facebook and Networking by Populist Leaders...

Reciprocity Network
In the entire network, there were 56 reciprocal connections, meaning that each of the three pages shared posts from pages that also shared them back. In Fig. 10, reciprocal connections were the ones marked by the red arrows. In Fig. 11, these were zoomed in to be seen more clearly. While previously it was shown that Boris Kollár had a bigger dissemination network (number of pages sharing his posts), the reciprocity was lower for him and his party then for OĽaNO. This means that there was more reciprocal promotion among OĽaNO party members than among WAF. Kollár seemed to endorse back very few of the people or pages that promoted him (only 11, compared to OĽaNO -29 and WAF -16). This might paint him as a more individualistic leader and his party as having weaker ties among its members, at least in terms of social media strategy.
In most cases, reciprocal connections were between party members / party pages and supporters like, for WAF and Boris Kollár: Lubos Hrica (supporter of the movement and of Salvini, living in Italy), Ludovit Goga (member of WAF), Sme Rodina -Detva (regional hub), etc., but also the Slovak Police page, interestingly enough. There was also cross-posting between all the three pages, and the link between them was also made by the dark humour satire FB page Zomri, which posted about OĽaNO and Boris Kollar, while both of them shared Zomri as well. There was also a direct link between Kollár and a prominent party member Milan Krajniak (MP, later minister of labour), as well as a curious connection with "sympathisers from Moravia" (region in the eastern part of the Czech Republic).
As for OĽaNO, the same pattern can be observed -namely reciprocity is mostly with party members or rather representatives (OĽaNO had for a long time only 4 members), as well as with the OĽaNO public group 18 . Yet while OĽaNO shared and was shared by WAF's leader -Boris Kollár, the same cannot be said about OĽaNO's leader -Igor Matovič, who shared posts by Kollár and WAF, but wasn't shared back by them. However, unlike WAF and Boris Kollár, OĽaNO also had reciprocal connections with different media channels (which might suggest more endorsement from media) like Dennik, Aktuality. sk (two of the most popular news sources for both political actors, as the first part of the research showed), and again Zomri satire page and Tablet.TV. Aktuality.sk is a professional news portal with investigative journalist news sources. Denník was an online only news portal that is no longer in operation. Tablet.TV is a video channel of the public service news agency. Since October 2020 it has changed its name to TASR TV (TASRTV.SK) and expanded its activities to live online broadcast both from regions and for commercial subjects. 18 OBYČAJNÍ ĽUDIA a nezávislé osobnosti -OĽANO, available at: https://www.facebook.com/groups/obycajniludia/. The public group has 6,426 members (August 2020) who are pretty active, posting around 20 posts daily. It was created in Jun 18, 2009 and changed it's name several times since 2018. The group is managed by the Facebook STOP Štátnej MAFIi (liked by 62,498 people).

Figure 11: Populist Pages' Reciprocity Network
Source: Marincea, 2020 Information Sources Shared on Facebook and Networking by Populist Leaders...

Network Centrality
As expected, there was a wide overlap between the FB pages that shared Boris Kollár and those that shared WAF (Fig.11 and Fig.12). Only 97 out of the 472 pages sharing WAF's posts shared only the party, and not the leader as well, while the rest -80% shared both. Most of these pages were either party branches or members, and some were also reciprocal connections (see above). But there were also pages like OĽaNO Sympatizanti (OľaNO´s Sympathisers) or ANTI-FICO a ANTISMER (two groups targeting negative sentiments towards the major political party and its leader in Slovakia at that time). Some signalled common concerns like Saving Slovakia, Active Citizens and For Sovereignty of Slovakia -all that could be seen as having populist overtones.
Interestingly enough, there was also quite some overlap with OĽaNO's public Facebook page (Fig. 10). 162 out of the 462 (35%) pages sharing OĽaNO also shared Boris Kollár at least once ( Fig. 12), and almost the same number was valid for WAF: 142 out of 462 -31% (Fig. 13). As could be expected, these pages largely overlapped and tended to be pages of party members or party branches. This shows that these types of pages, pertaining to the populist parties, play a central role in the dissemination network on social media.

Figure 12: Pages sharing 19 Boris Kollar and Sme Rodina
Source: Marincea, 2020 19 In the graph are shown only the pages that share each of the two pages at least 10 times. Source: Marincea, 2020 20 In the graph are shown only the pages that share each of the two pages at least 10 times 21 In the graph are shown only the pages that share each of the two pages at least 10 times Information Sources Shared on Facebook and Networking by Populist Leaders...
Nonetheless, there were a few FB pages that shared only OĽaNO and Boris Kollár (46 pages). Interesting overlaps seem to be two groups of expat Czechs and Slovaks in Switzerland and the Netherlands -supporters of these movements. There was -surprisingly -a minor overlap with Czech anti-populists, "We want better Czechia without Babiš and Zeman" and "We simply won't buy products produced by Babiš´s companies". Then it was relatively frequently shared "We should stop terrorists from Brussels." There was also an anti-fascist narrative -"We are fighters against fascism, Nazism and their ideologies!" This selection makes some sense, though. On the one hand, there were supporters of another party, "ZA ĽUDÍ" abroad" (a moderate, centrist party), and, on the other hand, a rather opposite group "KOTLEBOVCI -Ľudová strana Naše Slovensko (Oficiálna skupina)" (radical right-wing party with neo-fascist, mostly hidden, tendencies). There were some other connected expatriate groups (Tirol, Copenhagen, Ireland, UK, Bavaria, Czechia, Switzerland).
Among media sources, two of them were identified: alternative online magazine DAV DVAculture-politics revue, and less known FB discussion group Politika (Politics).
DAV DVA is self-defined as "civic initiative that promotes notions about alternatives towards the current economic-political system." 22 It is a follow up to interwar intellectual leftist group DAV and associated journal (1922)(1923)(1924)(1925)(1926)(1927)(1928)(1929)(1930)(1931)(1932)(1933)(1934)(1935)(1936)(1937). FB group Politika (Politics) had less than 400 members as of August 2020. It was created in January 2017. It was self-defined as "Free portal for political opinions and public issues opinions of narrower but somehow similar spectrum (for time being). Some (of our) opinions are strongly worded (with) others we would like to become familiar with." 23 Based on overview of key postings, it was pro-OĽaNO and against-Fico (former Prime Minister) group.
What can be said about the main connectors within the populist networks? The analysis found that there were only nine public pages that seemed to share all three profiles (Fig. 15), although with different frequency. These can be considered bridges between the different political actors and their different publics or some might be channels that either a) aim for more political pluralism and a somewhat more balanced position or b) pages that support one of the parties and frequently oppose the other, as a result.
They shared opposition towards Smer-SD which was until March 2020 the main party in government (Zomri, Politika, Antifico a Antismer, OĽaNO Sympatizanti, Za Jana a Martinu).
Analysis of shared pages by each of the three political actors showed some interesting patterns. First, only three of them shared all three FB pages a somewhat substantial number of times each (in this case we chose the threshold of 10 times each). Moreover, these were distributed very differently: OĽANO -Sympatizanti (Sympathisers) shared the OĽaNO page 250 times, which was to be expected, while it shared Boris Kollár only 15 times and WAF 12 times. This makes it obviously biased in favour of one of the two parties, and therefore in the second category listed above. This is to be expected from a group that explicitly contains party supporters.
Similarly less frequently shared, but overall much more balanced was the group Nevoliči  Source: Marincea, 2020 Somewhat more balanced was the satirical ZOMRY (13 shares Kollár, 10 shares WAF, and 19 OĽaNO). Obviously, political satire was popular among opposition parties and leaders.
Similar patterns became visible when we compared the number of times certain pages promoted only two of the competing political actors that shared each of them 25  Irrespective of the number of times each page is shared. 25 Keeping the threshold of min. 10 shares each Information Sources Shared on Facebook and Networking by Populist Leaders... further, more qualitative and in-depth exploration. What can be said is that OĽaNO supporters were more appreciative of the efforts of investigative journalists, while also being more focused against the government´s main party at the time. In contrast, Kollár was more associated with pro-Christian, national and social rhetoric.

Main promoters
We also analysed who the main promoters were for each of the three political actors. Considering that in order to ensure some level of visibility and consistency, each page should be shared several times, we kept the 10 times threshold for each. This resulted in 97 unique pages sharing Boris Kollár at least 10 times, 79 for WAF and only 24 for OLaNO. This suggests that the former two political actors had a wider network of "loyal" disseminators. In most cases, with very few exceptions, this network was made of pages of the party branches or members 26 . Other specific FB non-party pages were: For National, Christian and Social Slovakia) Za národné, kresťanské a sociálne Slovensko (43 shares), For Sovereignty of Slovakia, ZA SUVERENITU SLOVENSKA, Against NATO Bases, PROTI základniam NATO (27 shares) or Save Slovakia -Active Citizens, ZACHRÁŇME SLOVENSKO -AKTÍVNI OBČANIA (18 shares) among others. More or less the same pages shared WAF with a similar frequency.
OĽaNO followed a similar pattern of sharing. However, the reason for it being shared less overall is, most probably, that it had much fewer pages of party branches (or rather circles of supporters) of party members in its social network (indeed, OĽaNO had only 4 members in almost 10 years of its existence). This made it more diverse and, as we showed in the first part, stronger in terms of mutual support (reciprocity) or less centered on one figure.
Unlike their competitors, the top disseminators for OĽaNO were not other party members or branches per se, but the "fan" group OĽANO -Sympatizanti (250 shares) started by the STOP Štátnej MAFII page (Stop State Mafia -related to perceived state capture), followed by For Jan and Martina -a page honouring murdered investigative journalist Ján Kuciak and his fiance) Za Jána a Martinu (220 shares), We support Igor Matovič -Podporujeme IGORA MATOVIČA (209 shares), OBYČAJNÍ ĽUDIA a nezávislé osobnosti -OĽANO -the OĽaNO public group (182 shares), OĽaNO -Trenčiansky kraj -Trenčín region) (74 shares), Great Anti-Corruption March -Veľký protikorupčný pochod -sympatizanti (70 shares), ANTIFICO a ANTISMER (57 shares) etc. As identified earlier, these pages suggest key ideas that are behind political activism -protest against the government, especially the key governing party Smer-SD, represented by its leader Robert Fico, and believed to be responsible for the captured state.
It stands out from these results that, even though Boris Kollár clearly had a wider network on social media (FB), this was mostly due to the multitude of party-related FB pages and did not necessarily say much about real support from citizens or media institutions. In fact, the media did not seem to give him much visibility.
On the other hand, OĽaNO had a smaller party network, but more support from non-party sources, including citizens and media outlets -which refrained less from giving visibility to OĽaNO than to Kollár.

Conclusions
Clearly, in both cases digital sources were prioritised, and first of all FB content. This is hardly surprising considering the digital nature of communication via social media. As one could have expected, both FB pages prioritized party-related sources (mostly selected MPs) and in case of Kollár, European allies from the radical right (Salvini, IDP, MENF). However, it was somewhat surprising to find that both populist FB pages gave a slight preference to mainstream liberal newspapers and digital news websites (including of tabloid type), during all examined periods. We could identify "alternative" news sources only in very limited numbers. Among somehow by and large ignored media types sources one could identify both radio and TV channels. Instead, more often but still marginally were utilised videos produced by a public wire agency.
A similarity between the two political actors analysed was that they seemed to share information sources that leaned more towards the right side of the political spectrum. However, OĽaNO leaned more towards the conservatives, rather than populist or anti-immigration sources.
In absolute numbers, the role of public service media was very low, almost negligible, in the communication of both populist actors. There were only negligible differences in results for selected political periods. Fundamentally, neither populist actors showed any significant connection to alternative media sources. Both populist actors preferred a mixture of quality, tabloid and somewhat alternative media. The two populist actors have used their FB pages in all three analysed periods more as a self-promotion tool, to make themselves, their party and colleagues visible rather than to inform people on different issues. Nonetheless, it seems that Kollár used a higher variety of digital sources compared with OľaNO.
In terms of populist networks, Boris Kollár had the most intensive dissemination network 27 , while movements WAF and OĽaNO were very similar in terms of size of their network. There was a rather low reciprocity of cross-promotion. Kollár but also WAF endorsed back very few of the people or pages that promoted them (only 11, and 16 respectively), while OĽaNO endorsed 29 pages. In most cases, both Kollár's and OĽaNO's reciprocal connections were between party members or party pages and their supporters. However, unlike WAF and Boris Kollár, OĽaNO also had reciprocal connections with different media channels. Thus, OĽaNO was seen as more acceptable among the mainstream media.
To a certain degree, there was a connection to the emigrant-based support base as well as to the Czech -paradoxically -anti-Babiš (then P.M.) and anti-Zeman (then President) pages.
There were only nine public pages that seemed to share all three profiles. They shared the opposition towards the Smer-SD party -until March 2020 the main party in government, as well as 27 Kollár was shared 4721 times by 639 different public pages, while his party Sme Rodina -3399 times (by 472 unique Facebook pages) and OĽaNO -2985 (by 472 unique Facebook pages).