Norbert Merkovity[1]: Self-Surveillances in MPs Communication - How Parliamentary Representatives Use Facebook?* (FORVM, 2018/1., 147-162. o.)

I. Self-Surveillance on Facebook

The social networking site Facebook is an ambivalent phenomenon among politicians. On the one hand, they can ignore editors of traditional media outlets and can decide themselves what they wish to share directly with voters/followers. On the other hand, due to the nature of social media, it is not a one-way communication channel, even the slightest error could be noticed immediately; politicians must tolerate criticism in real time.

The philosophic expectations around social media were that it would create a wider public sphere, where the political actors could directly communicate with the public and vice versa. In fact, there is no sign of this versatile communication. According to previous findings on the politicians' and parties' use of internet, the political actors will use the web (and social media) as a tool for one-way-communication.[1] Basically, they share content with the followers, but do not wait for any response, which makes their communication style closer to a press conference. The politicians will surveil their own communication in order to control the interactions between themselves and the followers. While they are doing this, the press conference style of communication will end in 'self-broadcasting'. One can see that the internet does not create a more advanced 'public sphere' but rather

- 147/148 -

guarantee unidirectionality of communication in the world of politics.[2] I will use the term self-broadcasting of politicians as a consequence of self-surveillance. With the research of Members of Parliaments (MPs) I made a closer look on politicians' use of Facebook, thus we can see the self-broadcasting nature of their communication.

II. Politicians and the social media

The use of Facebook and Twitter social networking sites is now becoming common in politics. It varies from country to country which channels are considered a priority for politicians. The literature on Twitter is somewhat wider, because the structure of this social networking site allows easier access to examine the entries in scientific studies. During the research presented in this study the text was maximized at 140 words and the forms of interactions between the users were and still are regulated by the provisions of the platform. These specialties made Twitter easy to analyze by the researchers. We know from these researches that Twitter is not as popular as Facebook, but it is well suited to be the channel of opinion leaders.[3] This means that Twitter could even develop into the playground for and/or of politics.[4] For this reason, Twitter is not only suitable for politicians but also for political activists.[5] Thus, the analysis of ordinary citizens will be detectable in such research. However, it is more important in this particular study to analyze how politicians behave in the social networking space that some researchers consider as a strategic tool for public relations. Alex Frame and Gilles Brachotte compared French politicians' use of Twitter to corporate actors' use of social media and concluded that the platform is suitable to measure public opinion in real time, to follow and to react on current events and specific topics, to inform the public, to direct the attention and to make immediate communication with voters, journalists or other politicians and stakeholders.[6] Findings of Frame and Brachotte's research refer to high awareness of politicians. However, they made interviews with a small group of politicians; these results would be different if the sample was larger and instead of interviews another research method was used.

Instead of the interviews, I found it more informative to examine the content of politician's communication, because it indicated the nature of their shares. Noa Aharony's research is doing this. The researcher analyzed the content of shares by three leader's Twitter page and concluded that the communication style used by politicians on the Twitter channel was able to convey their messages, because they were not under the

- 148/149 -

surveillance of traditional media gatekeepers. "The results show that the three politicians made considerable use of Twitter as they released many tweets categorized as 'information about' and 'statements about', expressing their own independent opinions concerning various topics".[7] This means that politicians are using Twitter mainly for informational (self-broadcasting type of) communication, but not for interactive purposes. There is no reason to assume that it would be otherwise on Facebook. Sharing of information, opinions, and broadcasting will prevail on Facebook's less regulated environment. This and other trends will be demonstrated with the comparative examination of Facebook communication by MPs from 10 states.

III. Research method

I analyzed directly elected parliamentary representatives' (MPs) Facebook posts in the research. Facebook was chosen as the subject of analysis, because it is the most popular social network with 1.5 billion users worldwide at time of the research, and this fact did not avoid the attention of politicians as they are using this platform, too. I find it important to examine the directly elected Members of Parliaments, because Facebook is an important communication platform for re-election, and politicians could interact with voters not just during the campaigns, but between two elections, as well. Thus, indirectly elected or appointed MPs, typically representatives in the upper houses of parliaments, were not included in the research. The research analyzed MPs from 10 countries. These states are Australia, Austria, Canada, Croatia, Hungary, Ireland, Montenegro, New Zealand, Switzerland and the United Kingdom.

Altogether, five Westminster States - with the heritage of Westminster political system - were included in the research (Australia, Canada, Ireland, New Zealand and the United Kingdom) and five continental European states (Austria, Croatia, Hungary, Montenegro and Switzerland). Among the continental countries one can find European Union member and non-member states, old and new democracies, and each country is neighboring with at least one another country.

10% of MPs were examined from each country. The simple random sampling method was used to create the samples. Thus, every Member of Parliament had the same chance to be in the sample, the only requirement was that they had to have an active Facebook profile or page during the analyzed period. The requirement "to be active" means that the particular MP posted at least two entries during the reviewed period. The study did not want scale the MPs professionalization on Facebook, this is the reason that I did not make any difference between politicians who are using profiles and who are using pages. Some could say that pages indicate more professionalized use of Facebook, but some politicians are often using profiles in their professional Facebook communication. The difference of profiles and pages are more important to those users who want to have feedback on a wide range of followers' data. However, this data does not contain any information towards the user. This is the reason I decide to analyze both types of registration. Profiles, which were available for the public, were considered official Facebook outlets of the MPs. The

- 149/150 -

research assumed that if the politician would like to have private a Facebook account, then it would be closed for the public. Since the politicians are public figures, the open and available Facebook profiles were involved in the research.

Regarding the question of authorship, it often happens in politics that the representative has its own staff that takes care of her or his public communication. Sometimes the politicians acknowledge this, and the information is available on the opening page of their account, sometimes this could be understood from the SNS notes, and sometimes there is no direct evidence, but the tone and the style of the communication gives rise to suspicion. However, it cannot be said wholeheartedly that the absence of these signals means that the politician manages the account by her- or himself. I came to the conclusion that since the public speeches of politicians, which are typically written by a staff member or groups of staff members are considered as the 'politicians speech' it is not necessary to make a distinction between the accounts led by the politicians or by the persons from staff members. To conclude, the research project analyzed official Facebook accounts that were written under the name of politicians.

In compliance with these criteria, entries of 253 politicians from 10 countries were examined. The research period was three months, from November 1, 2012 to January 31, 2013. During the analyzed period a total of 15 121 Facebook post were written by the examined politicians. The posts were evaluated by quantitative content analysis.[8] They were classified into six categories, the first four of which are textual categories, thus, the entry could contain a picture or a link, but it was expected to have written content, too. The first four categories are private sphere, informational, subjective, and personal attack. The private sphere category is where the politicians wrote about their family, friends, personal happenings in their lives, but the seasonal greetings also ended up in this category. The informational category was used by MPs when they wrote about their work or directed the attention of their followers (e.g.: guiding the attention to an interview prior the broadcast or print, information on particular programme, openings, other official ceremonies or celebrations, etc.). The subjective category contains the representatives' personal opinions on events and/or institutions. The personal attack category is more explicit then the previous one. In contrast to the previous, it includes attacks which were directed against a person rather than against an institution or an event.

The two non-textual categories, when the politicians share a photo or link without any explanation, are link/video and photo. The link/video category means that the representative shared a hyperlink (video, audio, gif, etc. files) without any commentary. The photo category is typically a share of the MPs own pictures or albums without any explanation. I decided to use strict categorization, one entry could only belong to one category, which eliminated the possibility of analyzing the nuances in the tones of the posts, but in terms of research's goal the results will be clearer.

It is important to note that there were no election campaigns during the examined period in 10 countries. Therefore, the heightened communication state of election campaigns did not distort the results. The research managed to analyze the MPs during 'normal communication conditions', when the tone of political communication or rhetoric is not tuned up.

- 150/151 -

IV. MPs on Facebook

SPSS statistical software was used for processing the categorized entries. Thus, the research could highlight the significance of results, as well as on the standard deviation and standard error for a given country. The research data obtained from the study allows comparison between the countries. I chose the most transparent approach to introduce the results: each category will be analyzed separately together with countries' results. This approach does not detract the possibility of creating general outcome, but also allows detailed explanation.

1. Private sphere category

Introducing a certain level of private sphere of nationally or internationally known persons is the proof for global success of tabloid journalism. This wave also reached the world of politics, so the representatives will use this type of communication.[9] The real question is how typically providing representatives insights into their private lives. The research presented here gives partial answer. The results will show to what extent MPs use this category in analyzed states. Tables 1 and 2 show the results.

Table 1

Private sphere category in investigated samples

- 151/152 -

Table 2

Analysis of variance (private sphere category)

F (9.243)= 3.134 p=0.0015

The research found no real difference in the use of private sphere posts. The Canadian (5.5) representatives used this category frequently in the sample, while it was not used in Swiss sample (.0). However, the results of Montenegrin sample is out of line because the mean of 2.4 is reached with 1.7 standard error, which indicates that actually several politicians used this category, but they were very active. This could be contrasted with the balanced Australian sample where the mean is 2.8, the standard error is 0.5, while the standard deviation is 2.4. Hungary and Croatia is visible among the five continental European countries, but both are far below the Canadian mean. The Croatian representatives' mean is 3.467 while the Hungarian politicians' is 3.33. Naturally, since the Swiss sample had no private sphere posts, the differences between the samples are significant.

It can be stated regarding the private sphere category that it is less characteristic for Westminster countries, except for the Canadian sample. The continental European countries have much more uniformed representation in this category. However, the results show no real differences between the states' samples. Approximately 1-1.5 post entry is the average change between the countries.

2. Informational category

The informational category was expected to be the most popular, as it gives the most room for politicians to maneuver. Such notes are suitable to guide the follower's attention, but do not contain an opinion, which potentially could trigger hostility by the followers. In addition, these types of shares are under the influence of MPs, and thus, topics can be made public that traditional media channels would not broadcast. The selection and the publication of the information is in itself a result of a subjective process, there is no need to post too subjective entries. Politicians may influence their followers by informational type of communication. However, so far there is little evidence that online communication can influence the electorate's political behavior (Baumgartner and Morris, 2010; Kushin and Yamamoto, 2010).[10]

- 152/153 -

Table 3

Informational category in investigated samples

Table 4

Analysis of variance (informational category)

F (9.243)= 5.824 p<0.000001

Significant difference can be seen between the numbers of the informational posts. This category shows the greater differences between the individual state samples and group of countries, as well. The United Kingdom and Canada (again) should be mentioned here, since these two samples are high above to other country's samples, mean of 46 and 44.5. The means in other three Westminster countries are also high. Only the Hungarian sample (32.1) managed to reach the mean level of these five countries, it is even higher than in Australia and Ireland, and it is similar to New Zealand (32.2). However, the standard deviation and standard error is a bit larger than in New Zealand, what indicates that less Hungarian MPs wrote nearly as many entries. Altogether, the mean of Hungarian MPs' sample is more than 7.5 times higher than the Swiss sample, for example. This could be seen as a huge difference and shows that although the informational category is the most used category it is not so visible in every country. Austria, Croatia, Montenegro, and Switzerland are at the bottom of the list, which means that this category is not as popular in continental European states as in Westminster countries. Table 4 shows that the difference between groups' samples is enormously significant.

- 153/154 -

3. Subjective category

The subjective category, especially on Facebook, is suitable for MPs to express their own views and to make their standpoint clear on issues or to tell their own version of stories. In this way this category could be the manifestation of the phenomenon of media logic, if we accept the explanation of Lars Nord and Jesper Strömbäck that the media logic is a storytelling technique, in which one can discover simplification, personification, polarization, intensification, concretion, stereotyping, and enhancement. In circumstances where limited time is available to process the information (information overload) attention deficit could be observed, and there is increased competition among the various mediums for people's attention.[11] The representatives are on the other side of this phenomenon, but they are also fighting for the attention of the electorate. However, subjective communication could likely jeopardize their aim to securely provide information to the target audience. Thus, the informational communication will take over this role on Facebook in MPs posts. This is why it is expected that subjective category will be less popular than the informational category.

Table 5

Subjective category in investigated samples

- 154/155 -

Table 6

Analysis of variance (subjective category)

F (9.243)= 2.955 p=0.022

The high significance in table 6 already shows that the deviation between the results is lower than in the previous categories. The mean of the Austrian sample (35.5) shows that the country has another strong category since the informational mean was low. The second highest mean could be observed in the United Kingdom's sample, but this is misleading, because of the strong standard deviation. However, the deviation is high in Austria, too, but this is because of high number of data. According to this, the next sample will be Switzerland, where the subjective category is notable, too (13.1). It is worth mentioning that the subjective category is the most popular in the Swiss sample. Montenegro (0.4) and New Zealand (1.2) are on the other end of the list.

The subjective category is the most popular after the informational, but here the research found no real differences between Westminster and continental European states. The means of the Austrian and Swiss samples are strong, while the Montenegrin sample is on the other end of the list. Table 5 shows that if the samples with extreme results would be taken out the averages of means would vary between 5 and 10 in the remaining samples. This points to the moderate use of subjective category.

4. Personal attack category

Posts categorized under personal attack are those that have good chance to get into the traditional media stream because negative messages can catch the attention of wider audience.[12] However, using this style of communication could dispose followers who are not open to such communication. The representatives could unwittingly put a sharp opinion in such entries. The entries in this category may be able to build a personal brand, which could bring more followers on social media, but this effect is very limited.[13] Therefore, it was expected that the results will also support the expectation that politicians will avoid this type of communication.

- 155/156 -

Table 7

Personal attack category in investigated samples

Table 8

Analysis of variance (personal attack category)

F (9.243) = 1.344 p=0.215

Personal attack is the most neglected category in the samples. Austria and Switzerland should be mentioned here, where the subjective category was strong, but this did not end in high produce of personal attack (Table 7). Research has shown that the shares of opinions do not result offensive attitude and there will not be a significant number of attack-type entries. By contrast, the Australian sample, where the mean of the subjective category was the third lowest (6.3), produced the highest mean in personal attack category (2.1). Still, the mean is lower than it was in the previous category. Regarding the other samples, high values of standard errors show that a small number of MPs used this category, but those who did, used it intensively. Exceptions could be found in Australia, Canada, Hungary and the United Kingdom, where the differences are slightly greater between the values of means and errors.

The personal attack category is the first where no significant difference was found between the groups (table 8). 63 out from 253 representatives wrote posts in the samples which could be categorized under personal attack. Therefore, it has been shown that representatives are trying to avoid this style of communication. The research found no attacking posts in Montenegrin, New Zealand, and Swiss samples. If Australia is not counted, the most common mean value (after 0) was 1.

- 156/157 -

5. Link/video category

The first non-textual category is link/video. These records do not require much work from the politicians, because the posts in this category could be shared, for instance, by clicking on the appropriate button on news portals. Similarly to the informational category the entries of this category are suitable to guide followers' attention or to highlight important thoughts of others. But link/video is definitely less useful for politicians than the informational, because it does not interpret the news for followers. MPs do not use the option to create frames in order to guide the followers' conversation or thinking, although Jacob Groshek and Ahmed Al-Rawi states that successful online political and/or electoral campaigns show that the imposition of frames is very important for the candidates' general image.[14] Despite this, politicians shared large number of posts in this category.

Table 9

Link/video category in investigated samples

- 157/158 -

Table 10

Analysis of variance (link/video category)

F (9.243)= 4.734 p<0.00001

The results of link/video category are similar to subjective category. Not counting the results of lowest mean levels in Australian, Croatian, Montenegrin, and New Zealand samples, the records are between 4 and 9 in average. Hungary is an exception with an extremely high maximum (182) and the mean is much higher (23.5) than in any other sample. Therefore, it could be stated that this category is one of the most popular categories in the Hungarian sample. Ireland is the second on the list with a mean of 14.2, but the standard error is close to Hungarian samples level. The error is suggesting that compared to the Hungarian sample fewer MPs produced proportionally less link/video category in the Irish sample.

Comparing tables 9 and 10 shows another characteristic of this category. The samples are divided, namely, the representatives either share much of these posts or they share none. Therefore, discord dominates in the samples. This could also be seen at high significance level.

6. Photo category

The second non-textual category is the photo. In contrast to previous one, here representatives share their - mainly - own content, but do not attach any commentary to these images. However, these entries do not give room for interpretation, as well, and MPs do not ask for any further action from their followers. This category is different from the private sector's photo sharing, where the enterprises encourage costumers in social networking environment to comment, like or share their images in order to create community around the company or the brand. These costumers indirectly contribute to the promotion of the enterprises or the brand and while they are doing this they become ambassadors of the company.[15] If we look to the photo category from this perspective, then we can count missed opportunities for MPs to engage citizens.

- 158/159 -

Table 11

Photo category in investigated samples

Table 12

Analysis of variance (photo category)

F (9.244)= 3.620 p=0.0003

The results in table 11 are similar to link/video category's results, but the means are lower. Similarly to the previous two categories, differences between the Westminster and continental states could not be identified. The values of means are typically between 2 and 5. Photo category could be understood in the samples of Austria, Canada, Hungary, and New Zealand. For instance, the maximum is large in Hungarian sample (103), what is followed by Austria (77). Because of these reasons, the results show that the photo sharing is popular in these two countries' sample. In much weaker extent, but similar could be seen in Canadian and New Zealand sample, too. Regarding the rest of the samples, low numbers of these posts do not mean that politicians are not using photos in their communication. It rather means that the representatives usually add commentaries to their photo sharing and the images can be found in one of the other textual categories. However, this is not measured, since it was not the subject of the research. The significance level in table 12 - similarly to the previous category - shows that discord dominates in the samples.

- 159/160 -

V. From self-surveillance to attention guidance

It was expected from the results of research presented in this study to detect trends in MPs communication on Facebook, but the study also made visible the similarities between the analyzed country samples. I wanted to analyze what kinds of entries are shared by the MPs, how general is this if more countries are compared. Furthermore, is the presence of self-surveillance observable? If there is no sign of the self-surveillance, the MPs communication on Facebook should be very similar in each country. The research came to the conclusion that there are not so many differences between politicians on Facebook and on Twitter, Aharony's findings could work in this study, too.[16] The informational and opinion type of posts are present in MPs Facebook political communication, too. The closer look at the results will prove that - except for the personal attack category - although significant differences between the countries samples could be found, this do not allow us to draw general conclusions.

Only one category of the six is relevant to the differences between Westminster and continental states. The most popular, the informational category gives opportunity to highlight this difference. According to the results the research found some differences between the two groups of countries. The five samples of Westminster states use informational category a lot; they are guiding their followers' attention, while this is not typical for continental European states in the samples. One of the exceptions is Hungary, but the Hungarian MPs used non-textual categories in large numbers, too. The same could be said about the Irish sample (lowest mean of informational category and high mean of non-textual categories between Westminster countries), which is the other exception. These two countries' samples are not typical in their group of states.

The results of the research showed that representatives of Canada and Hungary are much more active on Facebook than others countries' MPs in the sample. These two had the highest values in most of the categories. The results indicate that those politicians who share links, videos or photos are doing it frequently. The results of the samples show that there is no middle way in these two non-textual categories.

Finally, it can be stated that especially the rituals of conflict in political communication are not so visible on Facebook. Online social media platforms are changing political communication, since these platforms operate in a distinctly different logic from the mass media (Klinger and Svensson, 2015: 1252).[17] Modern politicians are forced to use social media platforms as a communication tool during their work, in order to demonstrate their engagement with the issues what voters consider to be important. This will mean that MPs should identify with the public. While they are doing this the politicians will also self-surveil their own communication in order to get the highest attention from their followers. This communication style will mean a race for popularity: which profile will be followed by the public more? On the other hand, the parliamentary representatives could leave behind their wariness or 'political correctness' in their communication since there are no gatekeepers on social networking sites, they can self-broadcast their messages. This

- 160/161 -

communication style could be a tool for a new kind of mediatization, where representatives are able to set their own agenda bypassing the media. However, this attitude requires a new kind of logic, which is different from well-known (mass) media logic.

VI. Conclusion

Regarding the limitations of the study, I have to note that the research analyzed only a sample in each country. We cannot say what would have been the result if 100% of MPs Facebook penetration were analyzed. One cannot make any general conclusions from the results, this was not the aim of the research. Likewise, the outcome shows the communication nature of the representatives from the samples, but the results are valid only for the timeframe of the research. It is highly probable that the results would be slightly different if someone did the same study during election campaigns.

However, the trends could be seen from the results. For instance, the informational category is popular category among MPs. Informational posts are suitable to generate more followers and to direct their attention, but these messages are not containing opinions, therefore, they will not turn away massive number of Facebook users. Other trends are that politicians do not use Facebook to attack other persons and non-textual posts are used by a small, but active minority of MPs. The research concluded that these trends lead to a new form of political communication, that could be described through self-surveillance and guidance of follower's attention. Similar research on a different set of samples could come to a different conclusion, but as Aharony's study shows, the use of social media as a tool of one-way communication could be easily discovered from these studies. More emphasis should be put on the reasons behind the informational-like posts and the nature of these entries should be analyzed in the near future.

Összefoglalás - Merkovity Norbert: Önmegfigyelés a képviselők kommunikációjában: hogyan használják a Facebookot a parlamenti képviselők?

A kommunikáció, különösen a konfliktus és a konszenzus rítusai, amelyek a politikai kommunikáció alapintézményei, a korábbi pont-pont vagy pont-multipont közötti kommunikációból multipoint-multipoint kommunikációvá alakulnak át. Ulrike Klinger és Jakob Svensson azzal érvelnek, hogy az online közösségi média platformok megváltoztatták a politikai kommunikációt, mivel ezek a platformok a tömegtájékoztatási eszközöktől teljesen eltérő logikával működnek. Arra kényszerülnek a napjaink

- 161/162 -

politikusai, hogy kommunikációs eszközként használják a közösségi médiaplatformokat a munkájuk során, annak bemutatására (olykor bizonyítására), hogy elkötelezték magukat azokkal a kérdésekkel kapcsolatban, amelyeket a szavazók fontosnak tartanak. Ez annyit jelenti, hogy meg kell ismerniük, illetve be kell azonosítaniuk a saját közönségüket a képviselőknek. A nyilvánosságot arra használják a képviselők a kommunikációjukban, hogy minél több figyelmet szerezzenek, míg az újságírói szerep másodlagos lesz, szórakoztató és szkeptikus-cinikus, a jelentések értelmezőek, szubjektívek lesznek és a napirendet a politikai események határozzák meg.

A tanulmányban bemutatom a parlamenti képviselők Facebookon a nyilvánossággal folytatott kommunikációjukat, amit tartalomelemzéssel elemzek. 10 ország szerepel a kutatásban: Ausztrália, Ausztria, Horvátország, Írország, Kanada, Magyarország, Montenegró, Svájc, Új-Zéland és az Egyesült Királyság. A rekordok három hónap intervallumban kerültek gyűjtésre, 2012 novemberétől 2013 januárjáig. A minta a parlamenti képviselők 10 százalékát tartalmazza. Az országok összehasonlítása azon a tényen alapult, hogy a Facebook népszerűbb közösségi weboldal a kontinentális európai államokban, míg a westminsteri államokban a Twitter örvend nagyobb népszerűségnek. Összesen 15 121 Facebook-bejegyzés került kielemezésre. 6 kategóriába kerültek a bejegyzések. A kategorizált posztokat az SPSS statisztikai szoftver segítségével elemeztem ki. A projekt kezdetén azt feltételeztem, hogy a fent említett közösségi oldal használata arra irányul a politika világában, hogy a politikusok bemutassák az új IKT-k iránti nyitottságukat. Végeredményben azt kellett látni, hogy a politikusok valójában nem a választókkal való napi kapcsolattartás miatt használják a közösségi oldalt. A kutatási projekt korábbi szakaszában is hasonló eredmény született, akkor az e-mailek politikai használatának vizsgálata mutatott hasonló eredményt.

A tanulmány egyik felismerése, hogy az IKT-k és azon belük a Facebook közösségi oldal nem teremtett deliberatívabb "nyilvános szférát", hanem a kommunikáció egyirányúságát biztosítja. Ez az egy-multipont kommunikációs stílus biztosítja a vizsgált képviselők számára. Az egyik oldalról lehetővé válik a politikusok megfigyelése, ami a népszerű-ségük növelésében számít, azaz láthatóvá válik, hogy melyik profilt követi a nyilvánosság. A másik oldalról a képviselők a kommunikációjukban elhagyhatják az óvatosságukat vagy "politikai korrektségüket", hiszen nincsenek kapuőrök a közösségi oldalakon, önmaguk közvetíthetik üzenetüket. A megfigyelés így önmegfigyeléssé válik, ahol az egyirányú kommunikáció újfajta közvetítés eszközévé válhat, ahol a képviselők képesek középpontba állítani a saját témáikat. Klinger és Svensson szerint ez a fajta gondolkodás újfajta médialogikát követel meg, ami eltér a jól ismert (tömeg-) médialogikától. ■

JEGYZETEK

* The data collection for this publication was supported by Hungarian Scientific Research Fund - OTKA (No. PD 108908). The analysis and the Hungarian part of the research was supported by the project nr. EFOP-3.6.2-16-2017-00007, titled Aspects on the development of intelligent, sustainable and inclusive society: social, technological, innovation networks in employment and digital economy. The project has been supported by the European Union, co-financed by the European Social Fund and the budget of Hungary. The author thanks the students of University of Szeged and Bence Karvalics for their assistance in data collection.

[1] Merkovity, Norbert: Hungarian Party Websites and Parliamentary Elections. Central European Journal of Communication (4) 2011/2. 209-225. pp.; Merkovity, Norbert Hungarian MPs' Response Propensity to Emails. In: Solo, Ashu M. G. (ed.): Political Campaigning in the Information Age. IGI Global, Hershey, 2014. 305-317. pp.

[2] See: Imre, Robert - Owen, Stephen: Twitter-ised Revolution: Extending the Governance Empire. In: Bebawi, Saba - Bossio, Diana (eds.): Social Media and the Politics of Reportage: The 'Arab Spring'. Palgrave Macmillan, Sydney, 2014. 105-119. pp.

[3] Park, Chang Sup: Does Twitter Motivate Involvement in Politics? Tweeting, Opinion Leadership, and Political Engagement. Coputers in Human Behavior (29) 2013/4. 1641-1648. pp. Doi: https://doi.org/10.1016/j.chb.2013.01.044.

[4] Grant, Will J. et al.: Digital Dialogue? Australian Politicians' Use of the Social Network Tool Twitter. Australian Journal of Political Science (45) 2010/4. 579-604. pp. Doi: https://doi.org/10.1080/10361146.2010.517176.

[5] Bastos, Marco T. - Mercea, Dan: Serial Activists: Political Twitter Beyond Influentials and the Twittertariat. New Media and Society (18) 2016/10. 2359-2378. pp. Doi: https://doi.org/10.1177/1461444815584764.

[6] Frame, Alex - Brachotte, Gilles: Le tweet stratégique: Use of Twitter as a PR tool by French politicians. Public Relations Review (41) 2015/2. 278-287. pp. Doi: https://doi.org/10.1016/j.pubrev.2014.11.005.

[7] Aharony, Noa: Twitter Use by Three Political Leaders: An Exploratory Analysis. Online Information Review (36) 2012/4. 600. p. Doi: https://doi.org/10.1108/14684521211254086.

[8] Krippendorff, Klaus: Content Analysis: An Introduction to its Methodology. Sage Publications, London, 1980.

[9] Bingham, Adrian - Conboy, Martin: Tabloid Century The Popular Press in Britain, 1896 to the present. Peter Lang, Oxford, 2015; Ciaglia, Antonio - Mazzoni, Marco: Pop-politics in times of crisis. The Italian tabloid press during Mario Monti's government. European Journal of Communication (29) 2014/4. 449-464 pp. Doi: https://doi.org/10.1177/0267323114529535.

[10] Baumgartner, Jody C. - Morris, Jonathan S.: My FaceTube Politics: Social Networking Web Sites and Political Engagement of Young Adults. Social Science Computer Review (28) 2010/1. 24-44. pp. Doi: https://doi.org/10.1177/0894439309334325; Kushin, Matthew James - Yamamoto, Masahiro: Did Social Media Really Matter? College Students' Use of Online Media and Political Decision Making in the 2008 Election. Mass Communication and Society (13) 2010/5. 608-630. p. Doi: https://doi.org/10.1080/15205436.2010.516863.

[11] Nord, Lars W - Strömbäck, Jesper: Reporting More, Informing Less: A Comparison of the Swedish Media Coverage of September 11 and the Wars in Afghanistan and Iraq. Journalism (7) 2006/1. 85-110. pp. Doi: https://doi.org/10.1177/1464884906059429.

[12] Ceron, Andrea - d'Adda, Giovanna: E-campaigning on Twitter: The Effectiveness of Distributive Promises and Negative Campaign in the 2013 Italian Election. New Media & Society (18) 2016/9. 1935-1955. pp. Doi: https://doi.org/10.1177/1461444815571915.

[13] Lilleker, Darren G.: Interactivity and Branding, Public Political Communication as a Marketing Tool. Journal of Political Marketing (14) 2015/1-2. 111-128. p. Doi: https://doi.org/10.1080/15377857.2014.990841.

[14] Groshek, Jacob - Al-Rawi, Ahmed: Public Sentiment and Critical Framing in Social Media Content During the 2012 U.S. Presidential Campaign. Social Science Computer Review (31) 2013/5. 572. p. Doi: https://doi.org/10.1177/0894439313490401.

[15] van Doorn, Jenny et al.: Customer Engagement Behavior: Theoretical Foundations and Research Directions. Journal of Service Research (13) 2010/3. 253-266. p. Doi: https://doi.org/10.1177/1094670510375599.

[16] Aharony 2012, 600. p.

[17] Klinger, Ulrike - Svensson, Jakob: The Emergence of Network Media Logic in Political Communication: A Theoretical Approach. New Media & Society (17) 2015/8. 1241-1257. pp. Doi: https://doi.org/10.1177/1461444814522952.