Social media and other forms of communication technology have played a big role in the political formation of Egypt since the 2011 uprisings. In the wake of the uprisings, Egyptians called upon the new arsenal of social media technology in order to energise, organise and inform public protests. Social media did not cause the revolution, but it did provide a platform for people who were frustrated with the status quo to voice their opinions and gather support for coordinated protests. This finally led to the resignation of the president of Egypt, Muhammad Hosni El Sayed Mubarak, on 11 February 2011.

The influence of social media has, however, not dissipated since Mubarak's resignation. Debates on a multitude of political issues are constantly contested on various online social platforms. The social media activity surrounding the June 2012 Egyptian presidential elections serves as an example of how social media is still playing a determining role in Egypt. There are also those who believe that in the future social media can serve as an indicator, or even a predictor, of election outcomes. This paper examines the role of social media in Egypt's 2012 presidential election and assesses the potential of this form of media to serve as a tool for predicting the outcome of elections. In the interest of examining this possibility, this paper analyses sentiments gathered from the social media site, Twitter, before the Egyptian elections. The results from processing these sentiments show that the victory of Mohammed Morsi may have been predicted even before the second round of the election.

The organisational force of social media and the fall of Mubarak's regime

In early 2011, mass public uprisings engulfed Egypt. Motivated by the Tunisian revolution that led to the ousting of long-standing president, Zine El Abidine Ben Ali, millions of Egyptians took to the streets in protest of the Mubarak regime. The protests were precipitated by the lack of free and fair elections, the stringent state of emergency laws, corruption, police brutality and various other social and economic issues. Even though the protests were mostly of a peaceful nature, there were also violent clashes that caused the deaths of 846 people and injured a further 6,000.⁽²⁾ These casualties included Egyptian police officers, military personnel, medical staff and civilians alike. The biggest event in the revolution was the 18-day occupation of Tahrir Square by thousands of Egyptian protesters. The military attempted to set up civilian checkpoints to prevent protesters from entering the square, but this was unsuccessful. The protesters succeeded with the occupation and on 11 February 2011, Mubarak resigned from his position as president of Egypt.⁽³⁾

Young activists, such as Wael Ghonim, an Egyptian-born Google marketing executive, have played integral roles in the Egyptian revolution. Ghonim founded a Facebook page: 'We are all Khaled Said', in memory of the young Egyptian that was tortured to death by the police in Alexandria. Ghonim quickly gained mass support on his Facebook page and used his social media savvy in order to organise mass opposition against the Mubarak regime during the uprisings. On 14 January 2011, he asked his Facebook followers to take to the streets on 25 January 2011 and rise up against the Mubarak regime. Ghonim was secretly incarcerated on 27 January 2011 after coming to Egypt from the United States to support the protests. Bloggers such as Chris Di Bona and opposition figure Mostafa Alnagarand Habib Haddad, made it their personal task to advertise his mysterious disappearance and demand his release. Ghonim was consequently released on 7 February 2011 and subsequently became an even more powerful figure of the revolution. In a 60 Minutes interview on 14 February 2011,⁽⁴⁾ explained how the revolution works by comparing it to Wikipedia where everyone contributes. He also stated that the Egyptian Government shut down Facebook before the January 25 protests, but that the protesters just switched over to Google. This clearly shows the versatility and power of social media in the Egyptian revolution.⁽⁵⁾

The activists from the revolution are not resting today - political discussions on social media sites are still active in Egypt. Ghonim's Facebook page now has more than 2.3 million followers. The discussion on the page mainly focuses on the strong role that the military is playing in Egyptian politics since the 2011 uprisings.⁽⁶⁾ There have also been two tweet symposia (#TweetNadwa) held in Cairo. These symposia consist of discussion on the future of Egypt and take place online and offline simultaneously. Online participants ask questions and drive the discussion through Twitter and other social networks using the rules of Twitter. They are allowed to ask 140 character questions, and responses to these questions are limited to 140 seconds. At the first symposium on 12 June 2011, the role of the Muslim Brotherhood in Egypt's future was the point of discussion, while participants at the second symposium, held on 21 June 2012, looked at the original motivations of the 2011 revolution in order to ensure that it is not derailed.⁽⁷⁾ The new found role that social media plays in Egypt can still be witnessed in Egyptian politics and this is also the case for the 2012 Egyptian presidential election.

Mohammed Morsi emerges victorious from a two-round presidential election

After the resignation of Mubarak, the Supreme Council of the Armed Forces (SCAF) became the interim leaders of Egypt and promised presidential and parliamentary elections. During January 2012, parliamentary elections were held. The Freedom and Justice Party (FJP), which has strong links to the Muslim Brotherhood, won the parliamentary elections with 47% of all seats.⁽⁸⁾ Then, on 13 and 24 May 2012, the first round of the Egyptian presidential election was held. As there was no absolute winner in the first round, this was followed by a second round on 16 and 17 June 2012.⁽⁹⁾

The main candidates competing in the two rounds of elections from lowest to highest support were: Amr Moussa, an independent candidate, who gathered mass online support for his campaign. He received 11.13% of the votes in the first round of elections. The next candidate, also an independent, was Abdel Moneim Aboul Fotouh. He was expelled from the Muslim Brotherhood in 2011 when he refused to relinquish his independent political campaign.⁽¹⁰⁾ He received 17.47% of the vote in the first round of the election. Hamdeen Sabahi, who received 20.72% of the votes in the first round of the election, was the representative from the Dignity Party. After the first round he called for a suspension of the elections and stated that there were irregularities in the results that led to former Prime Minister, Ahmed Shafiq, progressing to the final round of the election.⁽¹¹⁾ The elections were not, however, suspended and Shafiq, with 23.66% of the votes, progressed to the second round. Shafiq ran as an independent candidate, but many tied him to the Mubarak regime as he was the former prime minister under Mubarak.⁽¹²⁾ The other candidate to progress to the second round of elections was Morsi, who received 24.78% of the votes. Morsi is from the FJP.⁽¹³⁾

In the final round of the election, the Muslim Brotherhood candidate, Morsi, competed against Shafiq, who, for many, represented the old regime and military power. This created a precarious situation for many Egyptians, especially Copts (Coptic Christians). Many Copts felt as if they did not really have a choice; either their president would be a remnant of the regime they fought to overthrow, or it would be a leader from the Muslim Brotherhood, who might suppress Coptic values. This led some Copts to refuse to vote in the second round of the election.⁽¹⁴⁾

On 24 June 2012, the election results for the second round of the presidential election in the Arab Republic of Egypt were announced. Mohamed Morsi from the Muslim Brotherhood, with a 51.73% ⁽¹⁵⁾ majority, beat his rival, Ahmed Shafiq, the former Prime Minister of Egypt, who garnered 48.27% of the vote. Morsi was subsequently sworn in as the first civilian President of Egypt. These results were celebrated by many, but there are also those, such as Joseph Fahim, a liberal Coptic Christian, who states that Morsi is "Not my president, and he'll never be."⁽¹⁶⁾ However, with such an important and close election, varying sentiments are to be expected.

The role of social media in the 2012 presidential election

During the June 2012 presidential election, social media played an important role, just as was the case in the populist uprisings of 2011. There has been constant dialogue regarding the election on various social media sites such as Twitter, Facebook, Youtube and even Wikipedia. According to Zynep Tufekci from the University of North Carolina, the Muslim Brotherhood especially maintained a sophisticated online presence and a positive narrative during and before the elections.⁽¹⁷⁾ This active online presence ensured a level of transparency that was previously unheard of in Egyptian elections.

The role that the Muslim Brotherhood played online can most clearly be seen in their online activities during the election night (17 June 2012). The Brotherhood posted members to all the polling stations. These members provided constant updates on the results, which were then relayed to the public on social media sites such as Twitter. This continued throughout the night until they were able to ascertain the victory of their candidate, Mohammed Morsi.

After this announcement the Brotherhood updated a Google spreadsheet, showing the results from all governorates of Egypt. The spreadsheet showed that Morsi won with a 4% margin. Tufekci states that the Shafiq camp was much more quiet on Twitter and other social media sites during and prior to the elections.⁽¹⁸⁾

When the official results of the election were finally released a week later, it showed that the Muslim Brotherhood's spreadsheet was accurate to within 0.03% of the official results. The official results were scrutinised and discussed in much detail on social media sites by academics, revolutionaries and dissidents alike. A well-known blogger, Mostafa Hussein, even ran a Benford's test on the official results, which would indicate if the results were manually manipulated.⁽¹⁹⁾ Tufekci argues that the availability of quick and accurate information that the Muslim Brotherhood provided the public would make it quite difficult for anyone to fix the numbers.⁽²⁰⁾ This availability of information and public discussion of the results on social networks therefore, according to her, leads to a much more transparent election than would be possible without modern technology.

The predictive capability of social media dialogue

A growing trend in social network analysis is to examine and analyse sentiments expressed on online social network sites such as Twitter and Facebook. The belief is that the computational analysis of large user-generated text archives can yield novel insights into the functioning of society.⁽²¹⁾ There are various goals for this type of analysis, such as measuring public support for certain issues or predicting future events ⁽²²⁾ such as the likelihood of uprisings, the behaviour of the stock market,⁽²³⁾ or even the outcome of presidential elections.⁽²⁴⁾

A study by VJ Um Amel illustrates how data from social media sites can be used to gain an alternative perspective on social events.⁽²⁵⁾ In this study, sentiments expressed on Twitter about candidates in the 2012 Egyptian presidential election are examined. This is done through automated sentiment and semantic analysis of Twitter tweets from 10 April 2012 to 24 May 2012. This study looks at certain important phrases (known as hashtags and preceded by the symbol #) in order to examine the discourse generated on specific issues and around specific candidates in the election. The study included both Arabic and English tweets. Furthermore, the study used an automated sentiment analysis programme in order to show whether expressions regarding the candidates were positive, neutral or negative.

The predictive capability of this type of study is still quite limited, according to Daniel Gayo-Avello from the University of Oviedo, Spain.⁽²⁶⁾ In his study, A Balanced Survey on Election Prediction using Twitter Data, he highlights a number of issues with predicting election results from Twitter data. He states that the simplistic sentiment analysis methods, the untrustworthy data from Twitter and problems with interpreting the data, make it impossible to be able to predict outcomes of elections from Twitter. This field of research is, however, still in its infancy and with time the quality of information from data analysis from social media sites might improve.

The fact that one cannot (yet) predict future events from social media data does not mean that one cannot gather useful insights from this data. The figure below shows the top five candidates (according to number of votes) in the 2012 Egyptian presidential election. Some of the findings of the Amel study have also been added. The column on average mentions represents the average number of mentions that each candidate received on a daily basis on Twitter during May 2012. The positive sentiment column represents the degree to which these mentions were positive, as a percentage. It is worth noting, as Amel points out in her article, that the former Prime Minister of Egypt, Shafiq, has the most mentions, but that many more of these mentions are negative, when compared with those mentioning Morsi.

Figure 1: Egyptian presidential election results and sentiment ⁽²⁷⁾

An argument can be put forward that negative sentiments expressed on Twitter, prior to an election, might indicate a negative correlation with the percentage of votes that a candidate will receive in the election. Or vice versa, that positive sentiments on Twitter might indicate a higher percentage of votes in an election.

Using data from the Amel article and the results of the first round of the presidential election, one can see a trend that seems to indicate a correlation between positive sentiment about a candidate on Twitter and the percentage of the vote that the candidate received in the election. From this simplistic, retrospective analysis of tweet data and presidential results, it would therefore seem that the above argument has at least some value for future studies.

Figure 2: Egyptian presidential election first round candidate sentiment analysis ⁽²⁸⁾

This is however a post-hoc analysis of data that the researcher knew would provide positive results. The concerns highlighted by Gayo-Avello ⁽²⁹⁾ still hold true and therefore there is much work to be done in order to provide accurate forecasts or predictions to elections, using Twitter data.

Concluding remarks

This paper highlights the continued role that social media plays in Egyptian politics. Groups, such as the Muslim Brotherhood, are successfully managing their online presence. They are engaging the online community not only to gain support, but also to ensure that the democratic political process is performed transparently.

The mass of social data that is constantly available on social media sites is also a resource that may play a growing role in the understanding of social events in the future. In particular, the second section of this article served to highlight the potential for social media analysis with regard to the predictive capabilities inherent in social data. It should, however, be noted that this form of analysis is still in its infancy and needs to be developed before it can provide truly meaningful results.

Written by Barend Lutz ⁽¹⁾

NOTES:

(1) Contact Barend Lutz through Consultancy Africa Intelligence's Election and Democracy Unit (elections.democracy@consultancyafrica.com)
(2) 'Egypt: Cairo's Tahrir Square fills with protesters', BBC News: Middle East, 8 July 2011, http://www.bbc.co.uk.
(3) 'Egypt crisis: President Hosni Mubarak resigns as leader', BBC News: Middle East, 12 February 2011, http://www.bbc.co.uk.
(4) 'Revolution 2.0',CBS News, 14 February 2011, http://www.cbsnews.com.
(5) Gordner, M.J., 'Whither post-revolutionary Egypt? The evolution of the Egyptian revolution and prospects for Arab/Muslim democracy', Consultancy Africa Intelligence, 2 March 2011, http://www.consultancyafrica.com.
(6) Aitamurto, T., 'How social media is keeping the Egyptian revolution alive', PBS, 13 September 2011, http://www.pbs.org.
(7) 'Update: Egypt's second #TweetNadwa', Aljazeera, 22 June 2012, http://stream.aljazeera.com.
(8) 'Muslim Brotherhood tops Egyptian poll result', Aljazeera, 22 January 2012, http://www.aljazeera.com.
(9) 'Muslim Brotherhood's Mohammed Morsi wins Egypt's presidential race', The Guardian, 24 June 2012, http://www.guardian.co.uk.
(10) Glain, S., 'Fault lines in Egypt's Muslim Brotherhood', The Nation, 12 September 2011, http://www.thenation.com.
(11) 'Egypt candidate to seek election suspension', Aljazeera, 27 May 2012,http://www.aljazeera.com.
(12) Fam, M., 'Egyptian Prime Minister Shafik resigns, caving into key protester demands',Bloomberg, 3 March 2011, http://www.bloomberg.com.
(13) Shehata, S., 'Copts between Shafiq and Morsi: An easy choice',Ahram Online, 12 June 2012, http://english.ahram.org.eg.
(14) Ibid.
(15) 2012 Egyptian Election Website, http://www.elections.eg.
(16) Fahim, J., 'Not my president', Egyptian Monocle, 2 July 2012, http://egyptmonocle.com.
(17) Tufekci, Z., '#Egypt's Muslim Brotherhood teaches a lesson in elections in the age of Twitter and Google Spreadsheets', Technosociology, 24 June 2012, http://technosociology.org.
(18) Ibid.
(19) 'Examining the election results with Benford's test', Moftasa.net, 21 June 2012, http://moftasa.net.
(20) Tufekci, Z., '#Egypt's Muslim Brotherhood teaches a lesson in elections in the age of Twitter and Google Spreadsheets', Technosociology, 24 June 2012, http://technosociology.org.
(21) Leetaru, K.H., 2011. Culturonomics 2.0: Forecasting large-scale human behavior using global news media tone in time and space. First Monday, 16(9).
(22) Woods, D., 'The predictive power of social media', Forbes, 27 July 2010, http://www.forbes.com.
(23) Bollen, J., Mao, H. and Zeng, X., 2010. Twitter mood predicts stock market. Journal of Computational Science, 2(1), pp. 1-8.
(24) Reed, T., 'Naval Academy study: Could Twitter predict presidential election?', Capital Gazette, 15 May 2012, http://www.capitalgazette.com.
(25) Amel, V.J.U., 'Egypt's presidential elections and Twitter talk', Jadaliyya, 26 May 2012, http://www.jadaliyya.com.
(26) Gayo-Avello, D., 'A balanced survey on election prediction using Twitter data', Department of Computer Science, University of Oviedo (Spain), 28 April 2012, http://arxiv.org.
(27) Compiled by the author using data available at: Amel, V.J.U., 'Egypt's presidential elections and Twitter talk', Jadaliyya, 26 May 2012, http://www.jadaliyya.com; 2012 Egyptian Election Website, http://www.elections.eg.
(28) Ibid.
(29) Gayo-Avello, D., 'A balanced survey on election prediction using Twitter data', Department of Computer Science, University of Oviedo (Spain), 28 April 2012, http://arxiv.org.