Media Sources Shared and Networking on Facebook. A Comparative Perspective1

Populist leaders tend to be more popular and more followed than their parties or movements. Exceptions, like Igor Matovič, or Jaroslaw Kaczynski, confirm the rule. The major differences in party versus leader ́s popularity („likes“) could be found for Recep Tayyip Erdoğan, Matteo Salvini, Alexis Tsipras and their respective parties. These three leaders were clearly FB stars (with caveat that Tsipras was actually not populist in his communication) and their parties seemed to be much less relevant for those who used FB. The most negative significant divergence in popularity on FB between a party and a party leader was noticed in the case of PiS and Jaroslaw Kaczyński. Kaczyński ́s FB page could be called as a niche phenomenon in Polish political communication. Also, Kaczyński was the least frequent actual user of FB among party leaders as well as the leader with the lowest popularity („likes“ in absolute and relative numbers) among political leaders in our sample among FB users. Similarly, FB seemed to be a rather irrelevant tool for PiS considering its FB popularity, although PiS actually communicated quite actively on this platform. While populists tend to be associated with alternative, highly biased, radical or conspiratorial media sources, the analyses in the national case studies showed that these types of sources were exceptions rather than the norm in almost all cases. The most often shared sources were digital sources or social networks. The least often shared were radio or TV channels. The rather ambiguous ideology promoted by Luigi di Maio and Boris Kollár was also reflected in their preferences for ideologically diverse media sources.

See e.g. Brussels, 30.9.2020, SWD(2020) 320 final, Country Chapter on the rule of law situation in Poland, https://eurlex.europa.eu/legal-content/EN/TXT/?qid=1602579986149&uri=CELEX%3A52020SC0320 EP, Press Releases, 14-07-2020 Rule of law in Poland: MEPs point to "overwhelming evidence" of breaches, https://www. europarl.europa.eu/news/en/press-room/20200712IPR83209/rule-of-law-in-poland-meps-point-to-overwhelming-evidence-of-breaches 3 Senior Analyst, Elcano Royal Institute; Senior Fellow, EU-Asia Institute at ESSCA School of Management and Co-Chair of the Political Economy Section, EUSA As can be seen in the Table 1, all selected populist parties showed rather high level of populism. On average, it reached 8.36 degrees on a 10 points scale. Occasionally, an argument could be made that we should have included political parties that were even more populist, such as more radical right wing rather than mostly typical populist parties selected here. However, primarily selection was done by local experts, and it took into consideration the participation in the European Parliament elections in 2019. Moreover, those more radical parties with a right wing authoritarian tendency either did not play an important role in politics, being ostracised to a certain degree by other populist or non-populist parties, as was the case of Kotlebovci -ĽSNS (Kotlebovci -People´s Party -Our Slovakia) in Slovakia, and/or their top representatives were sentenced for neo-fascist tendencies, specifically for running a criminal organization, such as as the Popular Association -Golden Dawn (XA) in Greece. In fact, the same happened to Marián Kotleba, the leader of Kotlebovci-ĽSNS, who was sentenced (subject to appeal) to jail for spreading fascists symbols.
Fundamentally, we attempted to make a distinction between populism and left or right (authoritarian) radicalism. This worked in most cases except France where the populist political spectrum is clearly and sharply divided into left-right dimensions. In other words, it is primarily defined by ideology rather than ideologically empty or populism. Similarly, the cases selected for Štúdie a analýzy Greece and Italy also resemble some left-right dimensions, but in more subtle ways. It should be noted here that in such cases it is questionable whether we discuss left-wing populist party or/and a leader (e.g. La France Insoumise -Rebellious France or Unbowed France (LFI) and Jean-Luc Mélenchon) or primarily populist subject. Following logical division, if there is prevailing and relatively coherent ideology, then priority in conceptualisation/terminology should be given to ideology, while populism is often just the way or style in which this ideology is expressed. Thus, LFI would rather be categorized primarily as radical left party and only secondarily defined as (more) populist.
Conversely, another case is that of parties or movements with no clear-cut ideology and/ or a mixed, contradictory or changing ideology, where policies are often unstable and rhetoric shows typical populist features such as anti-elitism, cherishing "the people" etc. In this case, then it is more likely that the political party or movement be categorized as a typical populist movement, often -especially in Central-Eastern Europe, an anti-corruption one. This is the case of Ordinary People and Independent Personalities (OĽaNO) and Igor Matovič in Slovakia.
As mentioned, selected Italian cases (M5S /Five Stars Movement/ led (then) by Luigi di Maio and the Lega (The League) led by Matteo Salvini) are more complicated cases for making such conceptualisation, although they are also showing some left-right distinctions. This discussion is important since, for example, Greek Syriza led by Alexis Tsipras showed high level of populism according to POPPA, but, in fact, the analysis presented in this volume on Greece (as well as previously cited analysis in that particular case study on Greece) showed that there was actually no populism in Tsipras' communication on FB. This finding raises a number of important questions such as -can we have a populist party with a non-populist leader, or at least one with a non-populist discourse strategy on social media? Is Tsipras only occasionally populist, or does he employ different communication tactics in different settings? Similarly, how can we understand Jaroslaw Kaczynski's reluctancy to communicate on FB? These questions cannot solely be answered in our present analysis, which can be seen as a limitation of the research, but they are directions for further inquiry in the study of populist communication and tactics.
Although our comparative approach does not focus on causes of populism, nonetheless, some of the case studies compared here revealed possible deeper salient issues or metapolitical questions causing populism to flourish. It is worthy to mention them here too -keeping in mind that these are advanced by the authors of the respective case studies, who usually found inspiration in works of others. Thus, in case of Poland (see case study), politics seems to be dominated by the question of who has the moral right to govern Poland, an issue most openly advocated by PiS. For Turkey (see case study), it is the socio-cultural divide from the past that pitted the ruling elites of the "center" against a culturally heterogeneous "periphery" or vice-versa. In other words, it is about feeling abandoned or ignored by the elite.
In case of UK (see case study), the populist upsurge then represented by UKIP reflected issues of sovereignty at the national level boosted by a key political decision on immigration. In other words, the metapolitical question was sovereignty as contemplated primarily, but not exclusively over the issue of migration. In case of France (see case study), the leftist LFI is probably an attempt to use an agonistic cleavage between 'the people' and 'the elite' in a radical individualised form of the French left. In contrast, for the right FN/RN, the hatred of foreigners and immigrants is the main motivation. This can be translated as a protectionist vision of society in both cases. For Slovakia (see case study), while OĽaNO can be seen as an anti-corruption movement, tolerant of minorities and to large degree benevolent towards migrants, right-wing movement We are a Family (WAF) is claiming to protect local population against potential migrants and other threats or sins or omissions (a lack of action or policy) committed by (then or previous) ruling elite. Yet the metapolitical question connecting both cases is the juxtaposition of a corrupt (in broad terms) versus a fair and just society.
In case of Greece (see case study), while long-term causes of populism were deeply rooted in poorly performing institutions at all levels, the rise of Syriza could be seen as a mirroring crisis of political representation (specifically, pro -and anti-austerity/memorandum policies), enhanced by the salience of the materialist cleavage. In case of Italy (see case study), 'refugee crisis' marks a crucial point in contemporary politics, giving rise to the Lega and Salvini at a national level, while M5S voiced primarily overall distrust towards political elites. Indeed, Italy is infamous with its instability of governments, thus showing a long-term crisis of representation of traditional parties. However, it should be mentioned again that populism in majority of these case studies has longer history -meaning also deeper roots. It appears that behind the rise of populism in all above mentioned cases lies a lack of ability or interest of mainstream actors in addressing timely and successfully national and/or supranational societal challenges (Kriesi 2015).

Populists and Facebook
Our research had two analytical parts. First, we explored sources shared by populist leaders and/or populist parties on FB. Second, we examined the network analysis of sources that shared populist leaders' and/or populist parties´ posts on FB. Before presenting this partial analysis, we show here some overview of populist parties/leaders' performance on FB to allow a reader to become familiar with general FB communication context in a comparative perspective. First, we show in the Table 2 dates when populist parties or leaders joined FB, and what was their popularity in terms of "likes" and "followers" in early March 2021. As can be seen from Table 2, first, populist leaders tend to be more popular and more followed than their parties or movements. Exceptions, like Igor Matovič, or Jaroslaw Kaczynski, confirm the rule. Unlike the others, Matovič had only a personal profile, which shapes the interactions differently, personalizing them even more. Jaroslaw Kaczynski also had as 'unofficial fanpage' only.
Second, in most cases, populist leaders created their FB profiles either at the same time as the FB profiles of their parties/movements were created, or, more often, earlier. This may suggest that the use of FB as a tactical political communication tool might have been adopted later, at least at an institutional level. Marine Le Pen is an exception, which can be explained by her role as successor of her father in leading the FN/RN. Similarly, the Polish exception -Jaroslaw Kaczynski´s late joining FB -can be explained by his personal negative attitude towards using social media actively 4 , and towards media in general (see Pacewicz 2021). This, however, did not prevent him, probably on advice of his team, to appear on TikTok in late 2020.
Third, the least frequent actual user of FB was Polish leader of PiS, Jaroslaw Kaczyński. His last entry on his FB dated January 24, 2021 (checked as of March 5, 2021). Moreover, it was a link to "High Class Racing". This finding show that a populist, or, more precisely in his case a national-conservative populist politician, may be successful in politics without having many "likes" or "followers" on social media. Electoral success is not always reflected in social media popularity, as the pool of the electorate may be more adept of using other media channels like television. Indeed, Kaczyński and PiS have secured mass-media coverage via the public service media which was captured since 2016 by way of different regulations increasing political control over them (Klimkiewicz, 2020). For this reason, social media coverage might not be as strategically relevant.
We also examined all FB pages to check the frequency and type of posting, as can be seen in Table 3.     As can be seen from Tables 3a-c, the most active users were the Italian Lega and Salvini, followed by the Polish Confederation. The least active -zero postings -proved to be Polish leader Jaroslaw Kaczynski. In contrast, his party, PiS, was actually above average active in communicating on FB.
The most frequent communication tool was (audio)visual, in the form of photos, followed by videos. Indeed, some reports suggest that ideologically conservative PiS is -ironically -the most progressive political body in using modern online communication strategies in Poland (Wanat 2019). Yet although there are PiS-aligned FB Pages and social networks in general that amplify content favorable to PiS, networks on the opposite side of the political spectrum disseminate counter-discourse (Bush, Gielewska, Kurzynski 2020). Furthermore, we were interested to compare share of "likes" within total population aged 15+ up to 65 years age. This can be seen as a rough indicator of importance of FB communication among our sample. There are minor differences between "likes" and "followers", but still, "likes" more likely reflect popularity -some may follow just for being informed. In fact, FB announced in January 2021 that it will do away with the "Like" button to instead focus on Followers (Perez 2021), but this measure has not yet been implemented at the beginning of September the same year. As can be seen from Table 4, the highest ratio of a party FB page "likes" among population was found in case of OĽaNO movement in Slovakia with 4,68%. In fact, OĽaNo could be seen rather as a virtual movement since it did not have relevant permanent party base -it had 4 founding members for a decade until it was forced by the law to open its ranks and accept a few dozen party members. Yet it was in government since 2020.

Media Sources Shared and Networking on Facebook. A Comparative Perspective
In case of the leaders, the FB winner was Recep Tayyip Erdoğan. His achievement was actually relevant both in relative and absolute terms. Simply put, Recep Tayyip Erdoğan is the FB man not only for Turkey, but apparently -in relative numbers -also globally. This seems to be in contrast with the emphasis put by local researchers on the use and importance of Twitter in Turkey. Alternatively, it can be that Twitter is even more important than FB for Turkish politics and that would make Turkish politics as social media driven exceptional case.
The lowest percentage of party/leader FB pages' "likes" was found in France for LFI and Jaroslaw Kaczynski in Poland. In other words, both LFI and Kaczynski are not really relevant FB entities. This was also true, to a certain degree, for PiS party. We have already mentioned that in case of Kaczynski his communication of FB seems to be rather of random or symbolic nature.
Furthermore, it is interesting to note the relative overlap of popularity of some parties' and leaders' FB pages and rather radical differences in popularity among other parties and leaders. For the former category, the closest overlap could be noticed for Sme rodina /We are a Family/ and its leader Boris Kollár from Slovakia. For the latter category, the most radical differences in party versus leader's popularity could be found for Recep Tayyip Erdoğan, Matteo Salvini, Alexis Tsipras and their respective parties. These three leaders are clearly FB stars (with the caveat that Tsipras was actually not populist in his communication) and their parties seem to be less relevant for those who use FB.
Before delving deeper into comparison of these results, we present some interesting findings of either expected versus unexpected results, both from political science and political communication perspectives.

Methodology
We employed a dual approach of the analysis, in order to uncover the types of information sources that were linked with populists -either because populists shared them, or because they shared the populist profiles. The analysis was carried out on Facebook data (Mancosu et al. 2020;Marincea 2020) from 17 Facebook public pages of populist leaders and parties from the 8 European countries. We selected three different time intervals totalling 13 months, with and without major events: before and after the European Parliament elections in 2019 (April -June 2019), during regular reporting (July 2019-February 2020), and at the beginning of the COVID-19 crisis (March-April 2020). The analysis was carried out at the following two levels: (1) Quantitative and qualitative content analysis of data sources (URLs). Using the Crowd-Tangle API (CrowdTangle Team, 2020), we downloaded all posts that shared a URL (news website or other types of websites, other Facebook pages, Twitter, YouTube etc.) from an average of 2 populist profiles in 7 countries: France, Greece, Italy, Poland, Slovakia, Turkey and UK. A simple codebook was elaborated for the classification of the extracted media sources (URLs) based on their type, ownership or political/ideological orientation. Based on this, we carried out a content analysis of the links extracted from the Facebook posts shared by the populist profiles during the analysed time intervals.
(2) Network analysis. Using the CrowdTangle API, we downloaded all references from the public pages that shared the links (URLs) posted by our sample of profiles in each country between 1 April 2019 and 1 June 2020 (13 Months total). This resulted in a dataset with 193,910 unique posts that shared the posts from the 16 profiles in the 7 countries analysed. We mapped these using NodeXL. The analysis is complimentary to the classification of sources shared by populists carried out at level (1) of the analysis and aims to reveal the networks around populist leaders, degree of reciprocity, the role of mainstream versus more alternative media sources, the density of connections, overlaps and differences.
This dual approach allowed us to map the wider network of interconnections around populist communication on Facebook. By following these two steps, we aimed towards two research objectives: (1) To identify the information sources that a) populists draw on and that b) populists promote, and to categorize them. The research questions that guided this enquiry were: RQ1. Do populists in different national contexts rely more on mainstream, traditional/established media sources, or do they prefer alternative news sources and social media, including citizen journalism?
(2) To provide an overview of the interconnections between populist actors and other relevant individual or collective public actors (i.e., media, politicians, celebrities, etc.). Mapping the populist networks allows us to see who are the main promoters of populist messages on social media, who are the amplifiers of populist messages, and to what extent the networks of different populist parties or leaders overlap. Hence, the research question driving the network analysis is: RQ2. Who are the main disseminators of populists' messages on social media and what degree of reciprocity is there between them and the populists they share?
This two-steps analysis is needed in order to capture a comprehensive picture of populists' connections on Facebook. When choosing this approach, the following arguments have been considered. Due to the access gained to the CrowdTangle tool developed by Facebook, we opted to use it because it allowed us to extract big data on both the links shared by populist profiles, as well as the profiles distributing these links and posts via the populist profiles. For step (1) of the analysis, we aimed for a more descriptive dimension of these sources and their categorization. The unit of analysis, therefore, was the URL domain. We developed a codebook (available on request due to space limitations) including categories such as: source type (i.e., digital sources, newspapers/magazines, radio, TV); type of source ownership -whether it is public (PSM) or non-public; scope of publishing (European/international, national or mixed, regional or local); type of printed or digital sources; political orientation of sources; whether the source is formally registered as media or not; and transparency of ownership. We decided to drop an initial variable related to the trustworthiness of the sources because of the difficulty in assessing it reliably.
After centralizing and aggregating the sources shared in each country by unique web domains or Facebook pages, the national experts proceeded to categorize them using the codebook. For this categorization existing secondary data was also used (reliable data concerning the bias or political positioning of different news outlets -ex. Reuters Institute Digital News Report). We decided to categorize as media sources Facebook, YouTube or other social media pages, considering that many posts on Facebook share other social media pages and that these platforms have become one of the main news providers online (Shearer & Mitchell, 2021). Leaving them out of the analysis would have given a very incomplete picture. In addition, this tendency in itself gives us a sense of the extent to which social media is becoming a source of information and competing news provider. Another way to establish the level of respectability and trust was considered to be through the creation of a new category: registered/non-registered media. In each country there are media that are registered as such by the profession and while in some this data is more readily transparent and available, in others this is less often the case. This lack of data accessibility can in and of itself be an indicator about the transparency and professionalization of the journalistic field. Another category was discussed -"controversial coverage/non-controversial" source, but was dropped because of its difficulty to operationalize reliably.
In the second phase -(2) the network analysis, a different dataset was used -one with external links, namely the public FB pages and groups that shared the 14 populists' posts. These were then mapped using the NodeXL software package developed by Microsoft Excel. For each country, the network contained one (Greece), two (most other countries) and up to three (Slovakia) different populist pages that the national experts considered most relevant to include. In many cases, these were political rivals, which made it all the more interesting to explore the degree of overlap or isolation of their respective individual networks.
The nodes in the network were represented by individual public FB pages or groups, and the edges -their connections with other pages. In order to have a manageable dimension of the network, we mapped only the edges of the main pages under study -the populist ones. Or simply put, who shared them, but not the connections between the pages that shared them as well. However, we did measure the reciprocity between the populist pages and those who shared them. This was done using the data from phase (1) -what populists shared. We also used the data from this first step when qualitatively looking into the sources that were most central in the network and those who were the main promoters of the populist profiles.
This dual approach allowed us to have more insight into the online sources that are close to populists in different countries and that may play a key role in reinforcing populist discourse and attitudes.
However, our research approach also has some important limitations. Probably the most significant one is that we were only able to identify those sources that were explicit from their URL. Populists can sometimes share videos made by themselves or from YouTube, which may actually be recordings of public or commercial TV channels. There is no way to identify these original sources of the recordings other than to go manually through each of these videos, which was not feasible as an overall strategy. Second, for many sources there is no information available regarding their registration status, ownership or political orientation, which limits the insights that we can have. And third, our analysis does not assess how the respective source is contextualized, if it is placed Štúdie a analýzy in a positive, neutral or critical context. For example, some pages may share the posts of populist leaders to criticize them. Such a positioning can only be identified either by manual coding -which is not feasible for thousands of posts, or through other automatic means, such as sentiment analysis -whose reliability is a matter of discussion. Our research does not cover these assessments. However, it is reasonable to assume that pages that are critical towards different sources will, in most cases, limit their distribution and visibility. As the analysis also shows, the sources that promote and give high visibility to a page are usually supporters of that page.

Expected Versus Unexpected Results
As mentioned, it was not expected that the Greek Prime Minister Alexis Tsipras from the unquestionably populist left party (according to POPPA index) would show no inclination to using populist rhetoric in his FB communication. Similarly, it was not expected that Polish populist right-wing leader Jaroslaw Kaczynski would -more or less -ignore FB communication.
Contrary to the expectation, there was no prevailing preference for alternative sources, understood in their negative connotation as disseminating hyper-partisan or fake news and hoaxes, often associated with populist parties or leaders (Hameleers, 2020;Corbu & Negrea-Busuioc, 2020;Waisbord, 2018). Although there were some occasional cases when populist leaders or populist parties shared or liked some non-mainstream publications (e.g. in the case of Kollár in Slovakia, or Le Pen in France), these were rather exceptions than the norm. The most typical example of a leader who used controversial sources was that of Italian leader Matteo Salvini.
At the party level, rather unexpectedly, Slovak OĽaNO had some reciprocity in media visibility with mainstream liberal media, in the sense that the party and its leader drew on liberal mainstream sources, which also covered the parties' activities.

Comparative Research Findings for Case Studies
First, we present in the Table 5 the basic types of media sources preferred or ignored by the populist parties and leaders. As can be seen in Table 5, the most often shared sources were digital sources (including social networks). The least often shared were radio or TV channels. Print and online versions of newspapers were usually the second most often shared source. Exceptions are Recep Tayyip Erdoğan and AKP -that shared just a very limited sample of sources which were pro-governmental TV stations.
Second, we were interested in the qualitative types of print media sources by and large preferred or ignored by the populist parties/leaders on FB. The results are presented in Table 6. Surprisingly, available data suggest that populist parties and populist leaders actually prefer non-tabloid (more or less mainstream) media sources. There are three partial or full exceptions. Slovak OĽaNO that preferred in majority of cases tabloid media. Then UKIP and N. Farage preferred almost equally both tabloid and non-tabloid media. Niche and other media could also be detected among about a half of the sample.   Furthermore, although populists can be seen located in both left and right, they are usually in "valence" version located in the centre. In other words, as put by Curini (2018), the content of political competition has substantially shifted from policy to non-policy factors, such as corruption issues. Thus, we were interested in finding their preference for media sources based on the media political-ideological orientation. The results can be seen in Table 7. It can be argued that using media sources that are difficult to categorize according to their political-ideological perspectives, can sometimes be indicative of populism. These include among our sample only Luigi di Maio and Boris Kollár. They have no typical ideology that would be reflected in their preference for ideologically close media sources. Alternatively, they are populists who are relatively open-minded in what concerns the media sources they consume or share. On the other hand, as discussed earlier, those populists who show strong inclination ideologically towards left or right media sources, can usually be seen as "radical left" or "radical right" respectively. These include Jean-Luc Mélenchon, Marine Le Pen, Alexis Tsipras and among parties Confederation.
Nigel Farage together with Matteo Salvini, and among parties PiS, UKIP and OĽaNO can be described as political entities that show -based on their sources preferences -a mixture between ideological and populist features.

Alternative/Conspiratorial/Controversial Sources Shared
We provide further a selected summary from the following country case studies. The Polish Confederation incidentally posted links to other websites which were opinion websites or blogs promoting radical right views or being in line with Russian propaganda like konserwatyzm.pl or kresy.pl or citizen journalism (for example, radical right medianarodowe.pl). For UK, among such sources were several rightwing commentators, including those associated with alternative right views. For Greek Tsipras, there were two links to songs uploaded on YouTube, which were unlike his other posts in both tone and content. For Slovak populists, there was no preference found for conspiratorial or radical sources either. In case of Italy, about one third of the external links shared by Salvini and Di Maio redirected towards the websites of mainstream newspapers.

Rarely linked or ignored media
In the case of Poland, among the media which were systematically ignored and very rarely shared were left and liberal media representing the mainstream of the public debate like newspaper Gazeta Wyborcza, weekly Polityka or TV channel TVN. For UK, none of the profiles shared citizen journalism sources, and only UKIP shared a very low number of official or expert sources. For Turkey, there was no variety as all media content was digital and produced either by the AKP media team or other accounts directly associated with the AKP. In fact, there were only three links that were not produced by the AKP media team. This shows how state monopoly has captured the online as well and especially social networks, which were some years ago still regarded as a potential counterweight to unchecked state power. For Slovakia, in absolute numbers, the role of public service media was very low, almost negligible. Remarkably almost absent during all periods were regional and local sources. For France, surprisingly, Jean-Luc Mélenchon ignored left-wing newspaper L'Humanité. But he referenced, instead, Le Figaro, Les Echos or Le Point, newspapers of the right or center-right.

Characteristics of the FB networks
In the case of Polish party Confederation there was some leaning towards more spontaneous grassroots fan groups or FB discussion groups which shared much more extreme content than the party itself. The two political profiles constituted separate spheres connected together through a very narrow number of FB pages. Reciprocity network of Alexis Tsipras was limited to posts shared between his account and the official account of his party, Syriza. There was no reciprocity between Tsipras and external, non-party accounts. Salvini's FB page network was far more extensive than Di Maio's. FB groups and pages sharing Di Maio's posts were rather limited in numbers and mainly organized as bottom-up initiatives by small groups of militants that were not formally linked to the 5SM. Conversely, those FB pages and groups more prone to share Salvini's contents tended to perceive themselves as local sections of the party. Slovak movements WAF and OĽaNO were very similar in terms of size of their networks.

Conclusion
There are useful lessons for researchers on this topic from a methodological point of view. First, the priority should be given to a lower number of researched items. For example, it proved to be less relevant to find whether a particular source was officially registered or not. Rather, it is political or ideological orientation, or other features (like business model -whether they are crowdfunded by audiences) that seem to be more interesting and relevant for analysis. Similarly, a summary on local debates on what constitutes "alternative/ controversial/ populist" media may be productive, too.
Second, within this context, for comparative studies, methodological and theoretical debates should be published separately (as we have attempted to do in this volume). This gives space for carrying out in-depth studies.
Third, it may be more useful to compare (traditional) electoral versus non-electoral periods rather than some exceptional circumstances like the pandemic crisis (save for martial law or when waging a war).
Fourth, research overview reflecting use of social media in a country is essential. However, this should not exclusively be done from the perspective of users (like statistics on usage of legacy and social media), but rather from the perspective of social media usage by political parties or other researched subjects (like in Tables 2 and 3 in this chapter). Most chapters in this volume have included a "research overview" that served such specific purpose.
Finally, for international comparative studies, focusing on general political-legal aspects (like freedom of the press or specific media regulation) and political party context is of outmost importance. Yet this should highlight "localised" available theories or hypotheses on the rise of populism among discussed political parties and leaders rather than a simple description of local situation.