Edval N. Zoto
European Journalism Association, Albania

Pre-elections PR in Albania: the use of influencing strategies on Facebook by the leaders of the major parties


This paper describes the communication strategies adopted by Albanian major political parties’ leaders on Facebook during the political campaign of the 2013 parliamentary election in Albania. Their strategies mirror two different approaches on public influence. The first approach, is elaborated by Malcolm Gladwell, an emphasises the role of extraordinary influential individuals in spreading ideas, while the second approach is presented by Duncan Watts, who challenges Gladwell’s theory, and sustains that the social structure, and how it is exploited, counts more than the simple engagement of super influential individuals for influence purposes. The strategy adopted by the eurosocialist Edi Rama, fits perfectly into the Watts hypothesis, and as the Socialist Party of Albania won the elections, it makes a case for elavuating the two influence hypotheses, but also in understanding and evaluating best practises in electoral campaings.


Key words: 2013 parliamentary elections, Albania, Facebook, Edi Rama, Sali Berisha



The 23rd of June 2013 was a special day for Albania. Parliamentary elections were held after a fierce political campaign. The two major contenders were the coalitions led by the Socialist Party (SP) headed by Edi Rama and the coalition led by the Democratic Party (DP) headed by Sali Berisha. The DP had been in power for the preceding eight years, during which period Albania had experienced regression not only in terms of employment and welfare but also in terms of law enforcement and democratic governance. Despite regression and governance problems, the DP was confident of obtaining a third term, while the SP had doubts as to the possibility of ascending to power alone. For this reason, the SP organized a coalition composed of numerous smaller political parties. 
Apart from the formation of coalitions, the influencing strategies conducted by the two main parties differed considerably. The DP concentrated its efforts on promoting the Government’s successes in the traditional media, which was flooded with pro-government clips, prime-time discussions held by pro-Government opinion-makers and journalists covering politics. The public meetings held by prominent members of the DP, almost invariably led personally by Prime Minister Berisha, were conducted in the form of massive meetings organized on public squares all over Albania. Berisha quite often referred to his mass of supporters during these meetings as “hurricanes”.
The SP, meanwhile, adopted a very different approach in publicizing its objectives and political programme. The entire squad of candidates, led by the party-leader himself, organized small but numerous face-to-face meetings in neighbourhoods and convention centres, as well as in rural areas. Although the SP also organized highly successful mass public meetings, the party became quite popular through its small roundtable-format meetings, referred to as “talk corners”.
Every single meeting was publicized by each of the SP candidates on social media websites by the posting of short descriptions and pictures of the events. Edi Rama himself led the way by presenting his agenda and the party’s objectives through his Facebook and Twitter accounts. The SP commenced its political campaign very early, almost a year prior to the election date, and by December 2012 its relatively “cheap” strategy was refined and was gaining as much attention as the “expensive” strategy of the party in power—a strategy which included political TV shows, movie clips, mediated Government meetings and Government advertising in the traditional media. Initially, Berisha had satirized Rama’s strategy of using Facebook and Twitter as a medium, but in late December he joined Facebook himself and directed all of his ministers to do the same. Within the space of only a few months Berisha’s Facebook account gained 500,000 subscribers, outdoing Rama, who had only 350,000 subscribers at the time and who had been active on Facebook for almost 2 years.i As elections day passed and the ballots were counted, surprisingly for all, including Rama himself, the opposition coalition won with more than 300,000 votes ahead of the DP coalition. The SP itself obtained over 180,000 votes more than the DP itself and the Rama Government took office.

Hypotheses on influence

In this paper I describe the communication strategies adopted by both Albanian leaders through their followers on Facebook and try to explain Rama’s victory from the perspective of the communication strategy on the social networking site. Before starting with the analysis, it is important to recognize two very important approaches towards the spread of ideas.  The first approach is Gladwell’s (2000) hypothesis of “influentials”. The second approach, which criticizes the former, is related to the “network structure hypothesis” proposed by Watts (2007).
Gladwell maintains that ideas spread like viruses, and that their distribution can be attributed to the role played by a few exceptional individuals. These individuals are identified as connectors, mavens, and salesman—individuals with persuasive skills who know a lot and have many connections. These special individuals directly affect the rapid spread of ideas among the wider public.  Other factors that help ideas spread with giant steps are the catchiness of the message and the specific context. Watts challenges this theory, claiming that Gladwell’s “influentials hypothesis” has deficiencies. For example, he mentions a well noted fact that Gladwell’s reasoning is based on anecdotal rather than empirical evidence. Watts maintains that ideas and messages are spread effectively when they find the right social network and right context. In other words, although “influentials” may help to spread ideas, the results achieved depend very much on the social network structure and the context within which they operate.
According to Gladwell’s hypothesis, either Albanian political leaders “tipped”, or their direct and personal influence grew extensively nationwide. Berisha’s political importance tipped when he joined Facebook, but also when he led massive meetings all around Albania. Rama tipped precisely on Election Day. Certainly, Rama and Berisha are highly influential individuals within Albanian society, and also on Facebook. In terms of “followers” on their respective profiles, they both have hundreds of thousands of subscribers. And if the Facebook numbers had matched the number of votes each won on the 23rd of June, Berisha would have clearly won the elections. But he didn’t. He also lost in some areas where he had previously dominated. Rama overcome in those areas and achieved a much unexpected victory. So what was going on there?

The research

Efaw (2009) recognizes the persuasive potential and cost-effectiveness of social networking sites, and especially of Facebook, which may also be exploited by political subjects to promote their ideas in order to gain supporters. Since the appearance of social networking sites, political forces throughout the world have created their spaces on these portals. In a very significant way, social media was used in the Obama’s 2012 Presidential Campaign. Rutledge (2013) analysed both Obama’s and Romney’s exploitation of social networking sites and identified Obama’s staff as more proficient in exploiting social networking sites. As she maintains, Obama-related postings dominated the social media because “his team got how networks work”. Further on, she asserts that “the real power of social media is not the number of posts or Tweets but in user engagement measured by content spreadability.” Delany (2011: 6) asserts that, in order to be effective in spreading political ideas that stick, online advocacy should be integrated with offline grassroots and effective press strategy.  The issue of which agents within a political force should embrace social media is clarified by Smith (2011). He mentions the necessity for local political candidates to maintain a presence on social networking sites rather than relying only on the presence of their national leaders (Smith 2011: 21–22). The research conducted by the aforementioned informed me in approaching both Berisha’s and Rama’s presence and activity on Facebook. Since both leaders were present on Facebook by the 5 December 2012, I collected data from each of their timeline postings from that date until the day before Election Day. I collected quantitative data, i.e., the number of “likes”, “comments” and “shares” for each posting, but also coded each posting in order to include them in one of the five categories described below:
Postings “on himself”- postings including pictures, ideas, personal narrations and/or of family, and friends not associated with the individual’s political activity.
Postings “on others” – postings that promote other political candidates (of the same party) or their views and ideas, and/or postings in the form of content sharing deriving from other political candidates.
Postings “on political issues”- Postings created by the profile owner about political programs and ideas, goals, elections-wise, propaganda etc.
Postings on “meetings with electorate” (shown as “Meetings 1”) – Postings showing and describing popular meetings, including pictures, text, videos etc.
Postings on “meetings with VIPs” (shown as “Meetings 2”) – Postings showing and describing meetings of the profile owner with very important people who may be foreign politicians, artists, etc. The nature of these meetings was not for the purpose of directly gaining more votes for the own political force.
As I put up the dataset with all the prescribed data, I built charts showing the performance over time in terms of quantity of “likes”, “comments” and “shares” (interactivity measures) for both Albanian political leaders. I also examined the trends of the interactivity measures and exposed their linear equations. By definition, the slopes of the equation determines whether the trend over time is positive or negative. Below are found the graphs of the performance of Berisha and Rama on Facebook in the specifiedperiod:

Figure 1. Berisha’s performance on Facebook, 5 December 2012 to 22 June 2013

Figure 2. Rama’s performance on Facebook, 5 December 2012 to 22 June2013

As shown in the graphs, the two leaders performed very differently on the social networking site. From the graphs, we can see that:
Rama was more active in terms of the quantity of postings. He posted 1,202 times during the whole period under examination, while Berisha posted only 519 times.
The performance of Rama’s interactivity measures was successful (all slopes of the trend lines are positive), while Berisha’s performance was unsuccessful (all slopes are negative).
In order to further examine their performance, I added up the interactivity measures values in order to obtain a singular measure of interactivity. By doing so I wanted to examine the highest possible values of interactivity over time. In other words I assume that each of the “likes”, “comments” and “shares” are derived from different individuals. (In reality, in most cases the same individuals liked, commented and shared the postings at the same time.) Below are shown the graphs built with the maximum interactivity values:


Figure 3. Berisha’s maximum interactivity trendline, 5 December 2012 to 22 June2013

Figure 4. Rama’s maximum interactivity trendline, 5 December 2012 to 22 June2013

The second set of graphs still identifies Rama’s performance as better than Berisha’s. Another fact noted through examination of the data is related to the “weight” of the interactivity measures value. Despite poor performance in the trend, Berisha’s values double Rama’s ones. I am not going to analyze this right away, but given the results of the elections, those numbers do not represent anything meaningful for my research. The interactivity measures values rank into several thousands, and they may as well represent hard-core supporters who did not fear any reprisals after openly supporting the party in power. The OSCE’s pre-election reports throughout the years of Albania’s transition mention the pressure exerted on citizens openly supporting opposition political parties, especially during Berisha’s rule. This pressure may have discouraged the opposition’s “Facebook community” in interacting openly with Rama through Facebook.
Also following Rutledge’s reasoning, I believe the trends of interaction are far more reliable in measuring influence than numbers without a context. In attempting to analyse the context of the Albanian leaders’ performance in interacting with their related electorate, I built separate graphs depicting each of the five contextual categories. The graphs below show the performance in terms of maximum interactivity over time for each of the categories coded:


Figure 5. Berisha’s interactivity trendline for postings “on himself”

Figure 6. Rama’s interactivity trendline for postings “on himself”

The graphs show the following results:
Rama’s postings “on himself” over the period in examination increased the interactivity in Rama’s profile, while Berisha’s postings performed in the opposite way.
Over time, Rama’s posted this category of postings in a different moment from Berisha.
Rama’s postings were more evenly distributed over time than Berisha’s.


Figure 7. Berisha’s interactivity trendline for postings “on political issues”

Figure 8. Rama’s interactivity trendline for postings “on political issues”

The graphs show that:
Both leaders successfully managed to achieve positive trends for this category of postings over time.
Rama’s postings weremore persistent and better distributed over time than Berisha’s. Also there is noted a high density of postings from Rama towards the end of the electoral campaign coming, while Berisha focused a little on political issues toward the end of his campaign.

Figure 9. Berisha’s interactivity trendline for postings “on meetings 1”

Figure 10. Rama’s interactivity trendline for postings “on meetings 1”

The graph shows the following:
Both leaders followed quite a similar pattern of density of postings of this category, especially during the last quarter of the campaign.
Even in this case, Rama achieved a positive trend of interaction while Berisha only a negative one.


Figure 11. Berisha’s interactivity trendline for postings “on meetings 2”

Figure 11. Rama’s interactivity trendline for postings “on meetings 2”


In this case, both graphs show a similar performance in terms of interactivity and distribution of postings over time by both leaders.
In relation to the category of postings “on others”, Berisha didn’t post anything that could be included in this category, while Rama did so. Below are the graphs showing Rama’s performance for this category:


 Figure 12 Rama’s interactivity trendline for postings “on meetings 2”

Notable in this graph is the fragment of concentration of these types of postings in Rama’s profile. The highest density of this type of postings is found in the second and third quarter of the campaign period. This also coincides with the period when Rama posted more about himself. As a result, it might be noted that Rama not only promoted other political figures but also promoted them alongside promoting himself. The same cannot be said for Berisha. Also, the trend for this category of postings is positive.
The following table gives a clearer image of the context-based postings of the Albanian leaders on Facebook:






On himself





On others





On political issues





Voters meetings





VIP meetings






519 Postings



1202 Postings

The following graphs offer an inclusive picture of all the postings, reflecting their contextual meaning and also the value of their maximum interaction measure, over time:
Figure 13. Berisha’s postings by type and interactivity value


Figure 14. Ramas’s postings by type and interactivity value

The stacked graphs reflect all the differences’ and similarities described above, but it is also notable that Berisha’s postings’ density inclined in the initial phase of the political campaign, while Rama’s postings become more numerous and more contextually distributed in the second, third, and fourth quarters of the political campaign.
What do all these results tell us? And how do they help explain Edi Rama’s victory over Sali Berisha?


After reviewing all the results from above, it is clear that Berisha’s strategy for influencing voters on Facebook reflected Gladwell’s suppositions. Being a very popular figure, after dominating the Albanian political arena for more than two decades, Berisha shaped his electoral strategy around himself. He is definitely an individual who may be identified at the same time as a maven, a connector, and a salesman. He constantly travelled all around Albania to lead his Party’s meetings where he presented the candidates running for Parliament. Those meetings were very popular indeed. His Facebook profile was also popular and he had far more subscribers than Rama. In addition, he also exploited traditional media influencing strategies, which consisted in transmitting continuously short clips promoting his successes, and also by organizing public debates in favour of his government. The individuals chosen to support the government were among the most famous journalists, opinion-makers, and celebrities in Albania. Various artists were engaged in his political campaign, especially pop-singers, in an attempt to attract the Albanian youth. The same cannot be said for his party’s candidates. Except for a few individuals, the candidates were all quite unknown to the public and they never reached a national audience. In every way, Berisha’s strategy was based on the Gladwell’s hypothesis of the “few”. The moment he joined Facebook, his account subscribers reached a very high numberincredibly quickly, leaving behind Rama’s two-year efforts on this perspective in as little as two months. In the Albanian network of Facebook, it was quite usual to see shared postings from Berisha’s profile.
Rama did not follow the same steps. He encouraged a network of his supporters to reach out to every small community in Albania. During the whole period he travelled probably as much as Berisha did, but his most characteristic meetings were more intimate, and often organized in the form of round-table discussions. Also, he pushed forward his local activists. Well-known candidates of the SP party were evenly distributed in all of the Albanian regions. Even on Facebook, while Berisha kept posting almost exclusively information about his meetings or Government successes, Rama would repost and share his supporters’ postings. Berisha’s leitmotif for the political campaign was the call “let’s go ahead!”, while Rama always emphasized the call “bring someone close to you with you on Election Day”. Going back to Facebook, Berisha’s most well-known candidates also mirrored their leader’sactions, while Rama’s large number of supporters were quite creative.
Rama’s campaign, knowingly or unknowingly, resembled more closely the Watts theory. Rama built and encouraged his network of supporters to spread political ideas and promises. He also became often a “router” rather than a source. Rama shaped his influencing campaign on Facebook by offering consistent information on political issues and meetings with his supporters. He was cautious in promoting himself, while he extensively promoted his candidates. The combination of the postings of different categories, and the meaningful presence, contributed to keeping the public engaged and increasingly interactive. As he promoted others continuously, he built trust for his whole network of candidates and local supporters. His influencing strategy was very different from Berisha’s. Berisha counted too much on himself when conveying his political ideas. He also publicized a lot the fact that he had more followers than Rama on Facebook, but apparently he failed to analyse his performance in terms of interactivity. While the number of his followers increased, his interactivity rates decreased. Thus he had more and more people following him, but not being interested enough in interacting with him.
Since Berisha lost the elections by a huge margin to Rama, it might be asserted that his influencing strategy also failed. From the other side, Rama’s strategy of successfully exploiting the social network and fitting into the context was a winning one.
In conclusion, what the Albanian leaders’ influencing campaigns on Facebook reveal is a major challenge to the hypothesis of “influentials”. Of course there may be idiosyncratic characteristics influencing the results of elections and the effectiveness of strategies, but still they provide more support for Watts’s claims about the importance of the social network structure and contextual challenges.


i According to data from Internet World Stats, there were nearly 1.1 million Facebook user profiles in Albania as of 2012. Retrieved at: http://www.internetworldstats.com/europa2.htm#al





Gladwell M (2000) The tipping point. New York: Little Brown Company.


Watts DJ (2007) Challenging the influentials hypothesis. WOMMA. Measuring the word of mouth. Volume 3, pp. 201–211.


Efaw JM (2009) Social networking services: The new influence frontier. IO Sphere, Winter Edition, pp. 1-7.


Routledge P (2013) How Obama won the social media battle in the 2012 Presidential campaign. The media psychology blog, http://mprcenter.org/blog/. Retrieved from: http://mprcenter.org/blog/2013/01/25/how-obama-won-the-social-media-battle-in-the-2012-presidential-campaign/; Date of access: September 10, 2013.


Delany C (2011) Online Politics 101: The tools and tactics of digital political advocacy.Version 2.0, Epolitics.com, January. Retrieved from: http://www.epolitics.com/download-online-politics-101/; Date of access: September 10, 2013.


Smith KN (2011) Social media and political campaigns.University of Tennessee Honors Thesis Projects. Retrieved from: http://trace.tennessee.edu/cgi/viewcontent.cgi?article=2442&context=utk_chanhonoproj;
Date of access: September 5, 2013.

Sali Berisha, with a copy of Gladwell’s ‘Outliers’on the desk of his office at home