Archive for July 17th, 2010

AFL team related Tweets by location

Posted by Laura on Saturday, 17 July, 2010

After I did A map of Team GWS on Facebook, which included a comment in the follow up about Twitter location, I wanted to do a similar map for other AFL teams using Twitter.

The problem is I can’t easily get all the follower data on Twitter in a way that is easy for me to work with. Instead, I used a tool called Searchtastic to get profile information about people who mentioned specific keywords in their Tweets. Searchtastic pulls that information from Twitter’s Search API and then outputs in Excel. Each Twitter search I did I related to a team: #gosaints for St. Kilda, Lions AFL for the Brisbane lions. (I tried to do at least three Twitter searches for each team to maximize my data set size.) The raw data used in this post can be found at AFLTweets.xls. After that was done, I went through the location field and determined the city, state, country for everyone who had listed a location. I then chose a few teams, added up how many references by city and created the map below. With the particular tool I’m using, I can only add about 7 or 8 teams before I can’t add more. The teams in this sample include: Adelaide Crows, Brisbane Lions, Carlton Blues, Collingwood Magpies, Geelong Cats, and St. Kilda Saints.



View full map

Things that strike me as interesting: No one from the Northern Territories and only one person from Tasmania and the ACT tweeted about the AFL. Is this because they have less internet access, fewer people, use social networks other than Twitter, or aren’t interested in AFL and these particular clubs? Western Australia also had tiny representation here with the max people per team from the city being one. The state has two teams but fans of the league in the state weren’t mentioning any of the competition. What also strikes me as interesting is that interest in the AFL in Queensland only goes as far north as Maroochydore. Beyond that, nothing. Brisbane suburbs are also largely absent. I should finish the rest of the teams and see if I can’t get a bigger data set to see if that effects the results at all.

Related Posts: