Micro-blogging sites, such as Twitter, seem to be all the rage these days. From Prime Minister Stephen Harper to my professor's cat, it seems as though everyone is twittering. The technology allows one to send out short (140 character) messages that are published on a website and broadcast to those within a person's social network. This can be done in a variety of ways, including using a computer or a cell phone's text messaging capabilities. It certainly has the potential to fundamentally alter the way in which we communicate, but how are people using the technology?

I signed up for a Twitter account last October and I was determined to figure out what all the fuss was about. One of the first things I did was sit down on election night and watch the stream of tweets posted to the #canadavotes hashtag. I still didn't get it. Many people were twittering the election results, but I could get the same information from the mainstream media and I was pretty sure that most Twitter users were just rebroadcasting the stuff they were seeing on television. So I tried insulting the Green Party, but not too many people took my bait. It didn't appear to be a forum for political debate, not to mention that the sheer volume of tweets flying by made it next to impossible to keep up with everything that was going on.

Fast forward a few months and I now have an established presence on Twitter and identi.ca, but I still find it hard to articulate how I use these services, or how those I follow use them. So I decided to look into the matter further to find out how people are using Twitter and what everyone on #canadavotes was saying that night.

As it turns out, people have come up with a number of different ways to categorize tweets. One study found that the intentions of Twitter users can be broken down as follows: Daily Chatter, Conversations, Sharing information/URLs, and Reporting news. Another paper lists the following top uses for Twitter: Staying in Touch, News, Quick Answers, Referrals, Awareness, and Promotions. While blogger John Walsh distinguishes six different types of tweets: The information Tweet, The wise Tweet, The introduction Tweet, The ReTweet, The status Tweet, The question Tweet, and The blog post Tweet. Presumably, people will use different types of tweets for different purposes. For my own analysis of #canadavotes, I used the following categories: conversational (@ replies), information/content, question, opinion, ReTweet, status, and statement (the results are displayed below).

As can be seen, a majority of the tweets were classified as opinion, followed by information tweets. The difference is that the former states an opinion on a topic or event and the latter gives information without making a judgment about it. A majority of the discussion centered around the election results (see the table below), which often fell into one of these two categories. Conversational tweets (replies to someone else's tweet) were also used about 22% of the time. While the 140 character limit of Twitter messages does not lend itself to making reasoned arguments, it does show that people are interacting with one another and not just sending out messages into the void of cyberspace.

As for what people were talking about, a majority of the discussion revolved around the results of the election (360 tweets), followed by the performance of the parties and party leaders (150 tweets), and issues relating to new media (101 tweets), which mainly centred around a discussion of Twitter and its effects on the political process. One very positive result of this analysis was the lack of personal attacks on other users, which is a common feature of many other forms of online communication (such as newsgroups and web-based discussion boards). Another common criticism of Internet forums is that they tend to be predominantly negative. How does Twitter compare? I coded each mention of a party or leader and whether or not they were portrayed in a favourable or unfavourable light (see the results below).

While there were more negative mentions than positive ones, a majority of the mentions were neutral. This is because people were sharing election results and not always commenting on them. Harper and the Conservatives took the brunt of the negative mentions, even though they took a plurality of the popular vote and won a minority government. Since this time, there have been attempts to organize conservative Twitter users, so it will be interesting to see if more Conservative supporters will be out next time around.

I also looked at how many people were tweeting and how many tweets each of them sent out (the results are displayed below). A total of 305 people tagged their tweets with #canadavotes between October 12, 2008 and December 5, 2008, with 94% of those posted on election night. Close to half the people posted a single tweet, while the other half posted between 2-20 tweets. Three users posted more than 20 tweets and no one posted over 80.

Have I sparked your interest? The full report is available here. You can also follow me on my Micro-blog.

Share Link: