Investigating Political Polarization on Twitter: A Canadian Perspective

During the 2011 Canadian Federal Election, a lot was written about the influence of social media on the election. On some level this is not very surprising. In just a short period of time, social media have altered many aspects of our daily lives, from how we teach and learn, to how we find and access health information (Fox & Jones, 2009). Now social media is also beginning to affect how elections and political discourse are conducted (Wattal, 2010; Gulati & Williams, 2010; Robertson, 2010; Chen & Smith, 2011). Politicians, party organizers, the media and the general public are moving in droves to promote and debate party platforms, solicit donations, organize political rallies or ‘flashmobs,’ recruit new supporters, and connect with other voters using social media. While this new reliance on social media has many obvious benefits, there is a growing concern that people are becoming more politically polarized on social media. This is especially prevalent among supporters of opposing political parties or ideologies.

Next Thursday (September 20, 2012), I will be presenting the preliminary results of one of our Lab’s ongoing research initiatives, “Political Polarization on Social Media” at the “Internet, Politics, Policy 2012” Conference. The conference is being organized by the by the Oxford Internet Institute (University of Oxford) for the Journal of Policy and Internet, in collaboration with the European Consortium of Political Research (ECPR) standing group on Internet and Politics.

The paper, “Investigating Political Polarization on Twitter: A Canadian Perspective” investigates whether or not political polarization exists in social media by using social network analysis to analyze a sample of Twitter messages posted by users on, a popular microblogging platform, during the 2011 Canadian Federal Election. The preliminary results suggest that there are some pockets of political polarization on Twitter, but at the same time there is evidence that as a communication and social networking platform, Twitter may be able to facilitate open, cross-party, and cross-ideological discourse.

Figure 1: “Who mentions whom” Twitter network indicating types of accounts based on party affiliations.

Note: The nodes in the network have been re-arranged and grouped based on shared political views. Such visualization demonstrates the presence of cross-ideological connections between people.

The full text of the paper is available online at

Gruzd, A. (2012). Investigating Political Polarization on Twitter: A Canadian Perspective. Internet, Politics, Policy 2012: Big Data, Big Challenges: Internet, Politics, Policy 2012: Big Data, Big Challenges? September 20–21, 2012, University of Oxford, Oxford, UK.


Wattal, S., Schuff, D., Mandviwalla, M., and Williams, C. B. (2010). Web 2.0 and politics: The 2008 U.S. presidential election and an E-politics research agenda. MIS Q, 34(4), 669–688.

Girish J. Gulatia* & Christine B. Williamsa (2010). Congressional candidates’ use of YouTube in 2008: Its frequency and rationale Journal of Information Technology & Politics, 7 (2-3) DOI: 10.1080/19331681003748958

Robertson, S. P., Vatrapu, R. K., and Medina, R. (2010). Online video “friends” social networking: Overlapping online public spheres in the 2008 U.S. Presidential election. Journal of Information Technology & Politics, 7(2-3), 182-201.

Chen, P. J., and Smith, P. J. (2011). Digital media in the 2008 Canadian election. Journal of Information Technology & Politics, 8(4), 399-417.

On the Media. (2011, June 17). The echo chamber revisited: Transcript. Retrieved from

Young, J.R. (2010, July 22). How social networking helps teaching (and worries some professors). The Chronicle. Retrieved from