![Sample our Politics & International Relations journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5947/TOC-Subject-Sample-2021-Politics-International-Relations.jpg"})
ABSTRACT
This article shows how the presence of an election campaign affects the way in which the main political leaders of 18 countries communicate on Facebook. Findings indicate the presence of robust qualitative and quantitative differences between campaign and peace-time periods. More specifically, this comparative analysis points out that election campaigns have a positive effect on the frequency with which political leaders update their Facebook pages and on the personalization of political communication, while a negative effect is associated with the publication of posts about policy issues and with a negative rhetorical strategy. No effect has been discovered concerning the privatization of political communication on Facebook. Finally, independent t-test analyses and bivariate correlations denote a generalized process of homogenization, whereby there is no single main impulse which governs the different communicative choices taken by political leaders on Facebook as an Election Day approaches.
Notes
1. See the cases of Mario Monti (Italy) and Ed Miliband (UK), who have hired one of Barack Obama’s heads of communication for their failed campaign in 2013 and 2015.
2. Considering the analysis of how political actors communicate on Facebook, so far the literature does not offer more than case studies or comparisons of two cases (see e.g., Bronstein, Citation2013; Larsson & Kalsnes, Citation2014; Parmeggiani, Citation2015; Samuel-Azran, Yarchi, & Wolfsfeld, Citation2015).
5. The 2012 Dutch and Venezuelan elections, which took place on 12 September 2012 and 7 October 2012, respectively, have been excluded by the analysis since they happened less than 60 days after the first day taken into consideration by this comparative research.
6. Only countries of more than three million inhabitants have been taken into consideration for each political geographical area included into the analysis. To use numeric thresholds arbitrarily is not an unusual operation in comparative research (e.g., Lijphart, Citation1999: 53).
7. The main political leaders of Australia and New Zealand have been included into the analysis since these two countries are part of the so-called “Anglo-Saxon area” (Morlino, Citation2005: 56).
8. See, for example, the case of Gutierrez (Ecuador) who did not overcome the 15% threshold in the 2013 presidential election, while he did overcome this threshold in the previous elections, which took place on 29 April 2009. This is the reason why Gutierrez has been included in the data set from 1 September 2012–17 February 2013.
9. A party leader can be replaced after a party meeting, following his/her resignation, etc. At the same time, a prime minister or a president can lose his/her executive position following his/her resignation or after having lost an election while he/she was in office. Consequently, some cases have been included into the data set only in a subset of the 26 months taken into consideration. This is the case, for example, of Mario Monti, Italian prime minister from 16 November 2011–28 April 2013, who has therefore been included in the analysis from 1 September 2012–28 April 2013.
10. The guiding criteria applied to the sampling process are the following: for each leader, at least one post every 2 days has been analyzed, at least one post per day when the leader is involved in an election campaign. An hypothetical leader A has been included in the analysis for 400 days and he/she has published 2000 posts in this time period. At least 230 posts (randomly chosen) have been coded in this third case. More precisely, 170 posts were coded during the 340 days in which the leader was not involved in an election campaign, plus at least 60 further posts during the 60 days before the corresponding Election Day. First, at least 60 posts related to the 60 days before the corresponding Election Day have been randomly chosen and coded. Every other post has been coded if this hypothetical leader A has published, for example, 100 posts in these 60 days. Then, a further 170 posts have been randomly chosen among the 1,900 remaining posts included in the data set but published outside the 60 days before the corresponding Election Day. The choice to double, if necessary, the number of posts coded during the 60 days before an Election Day, follows the empirical observation of the data. With the unique exception of Werner Faymann (Austria), all the leaders included in the data set raised the number of posts published in their Facebook pages significantly during the election campaigns. To code only one post every 2 days even during election campaign would have impoverished the empirical analysis too much, underestimating this specific political time period.
11. A leader has been considered as populist when the literature defines her/him or his/her party as such. Four Western leaders, Beppe Grillo, Silvio Berlusconi (Italy), Strache (Austria) and Siv Jensen (Norway) (Bordignon & Ceccarini, Citation2013; Fryklund, Citation2013; Heinisch, Citation2008; Tarchi, Citation2008), and two Latin-American leaders, Rafael Correa and Lucio Gutierrez (Ecuador) (Manrique, Citation2013; Peruzzotti, Citation2008; Weyland, Citation2013) have been coded as populist by following this criterion. None of the other leaders included in the data set has been coded as populist since no significant references in the literature arguing for labeling these leaders as populist has been discovered.
12. To divide the data set into left-wing and non-left-wing leaders the following four references have been used: Castles and Mair (Citation1984), Bakker et al. (Citation2015), Kitschelt, Hawkins, Luna, Rosas, and Zechmeister (Citation2010) and Bordignon and Ceccarini (Citation2013).
13. Campos (Brazil), Matthei (Chile), Romney (USA) and Rudd (Australia) have been excluded by the analysis because the number of days in which these leaders were not involved in an election campaign were too few in order to conduct a significant comparison with the period in which these leaders were involved into an election campaign. Morales (Bolivia), De Wever (Belgium) and Vasquez (Uruguay) are excluded as well by this analysis, since they have never opened a personal Facebook page. Finally, even Maduro (Venezuela) and Spindelegger (Austria) are not part of this analysis. The former because he did not use his Facebook page at all during the 2013 presidential campaign in Venezuela; the latter because he deleted his Facebook page during the time period taken into consideration in this comparative research.
14. Most of the leaders communicate on Facebook in the 24 hours before the Election Day, even though an electoral silent law is present. For this reason, the time period taken into consideration is 61 days, since the Election Day is included in the analysis.
Additional information
Notes on contributors
Diego Ceccobelli
Diego Ceccobelli is a postdoctoral research fellow at the Institute of Humanities and Social Sciences of the Scuola Normale Superiore. He studies political communication in comparative perspective, with a particular focus on social media.