Tweet, Tweet: What Haligonians Are Saying About the 2012 HFX Election #HRMVotes #voteHRM #hfxpoli

In the week leading up to this year 2012 Halifax Regional Municipality municipal election, the Dalhousie Social Media Lab has been collecting public tweets about the election. This project is part of the Social Media Lab’s initiative to provide independent social media coverage and analysis of important public events. Our findings are made available here on the Lab’s website, on Twitter and shared with mass media outlets such as CBC Radio and others.
In the photo (from top left corner clockwise): Madeline Driscoll (@MadCDriscoll), Anatoliy Gruzd (@gruzd), Ryan Dyck (@ryanjordandyck), Philip Mai (@phmai), Catherine McGoveran (@kittmcg)

Since Monday, 13 October, 2012, we have collected a dataset containing over 4,000 Twitter messages (tweets) from 813 twitter users who tweeted using one of the three popular hashtags that have emerged in this election cycle: #HRMVotes  #voteHRM  #hfxpoli. To capture these tweets and analyze what people have been tweeting about in relation to this election, we used a tool designed by Anatoliy Gruzd(@gruzd), the director of the Social Media Lab. The tool is called Netlytic, a web-based system for automated text analysis and the discovery of social networks from electronic communication such as Twitter, online forums, blogs, and chats. Among other things, Netyltic can automatically tell us what words, accounts, and hashtags are being used and mentioned in the tweets from our dataset.

This image shows the top fifty most-often mentioned words along with the actual number of times a particular word was used in tweets about the election. In the remainder of this post, we will explore why and in what context the top five words in this tag cloud were mentioned.

@tomaskformore (727) / #martin (378)
It is clear from the tag cloud that @tomaskformore, which is the Twitter handle for Tom Martin, a mayoral candidate in Halifax Regional Municipality, was mentioned the most, seven hundred and twenty seven times to be exact. This fact is interesting in light of the fact that Mike Savage, Tom Martin’s leading challenger for mayor, has 4 times more followers (3628)on Twitter, while Tom Martin Twitter account only had  905 followers. The reason why Tom Martin was mentioned the most will be discussed in more detail in a separate post.

#vote (648) / voting (243)
The next most often used word in tweets about this election is the word “vote”. This word was generally being used in a positive manner in this election. “… don’t forget to vote“, “… #voteHRM“, “… I just voted” are commonly occurring phrases associated with this term. Clearly, HRM citizens are very civic minded and encouraging their fellow Haligonians to exercise their democratic right to vote.

online (246)
The word “online” is the third most often used word. It was mentioned in reference to the new e-voting method. Many tweets including this word were tweets encouraging others to vote online, mentioning how fast online voting is, or declaring that they just voted online. E-voting is becoming increasingly common and popular, with 22.47% of HRM voters making use of this voting method for today’s municipal election.

election (267)
The term “election” in this tagcloud is different from some of the other top terms, in so far as it is associated with a variety of topics. Twitter users are discussing particular issues, how and where people can get election coverage and information, and counting down to election day. The scope of this term is much broader than the others and the content it brings together is interesting to follow.

@teamsavagehrm (257)
Team Savage made a showing on Twitter as part of the top five most often mentioned word in tweets about the election. While the campaign Twitter presence has not been as strong as Tom Martin’s, it will be interesting to see the impact these two teams has on translating online discussion and activity into votes (especially among younger voters). Social media activity cannot predict results, but it will be interesting to see in the aftermath if/how it may have had an impact.

Below is another way to view this data is to look at it over time, as collected in the days leading up to the election. This shows how the conversation has evolved over time, and when different words were used and topics were discussed.

While the above image shows how topics and words have been used over time, this chart below shows how the discussion has been building in the days leading up to election day. Users are getting increasingly involved using the above words and sharing messages about today’s election.

More blog post about this election will be forthcoming.

Written by Catherine McGoveran (@kittmcg) and Philip Mai (@phmai), with contribution from Anatoliy Gruzd(@gruzd)