RESEARCHERS AT the Oxford Internet Institute have been exploring what geotagged tweets can reveal about the offline world. The research has led them to create a Twitter map to try and understand the link between the club football fans support and the places where they live.
The data included all geocoded tweets mentioning any of the Premier League football teams and their associated hashtags, (for example #MUFC) that were sent between 18th August and 19th December 2012. The tweets were then matched to postal codes in order to map a fairly detailed geography of results.
The data shows that contrary to popular belief Manchester City is not the “real” local team and that Mancunians support both the city team and Manchester United. Only a few parts of Greater Manchester showed significantly more tweets mentioning Manchester City over their local rivals.
The team at the Oxford Institute for Internet were also able to map the Merseyside rivalry between Liverpool and Everton. The twitter map shows that while Liverpool is slightly more popular in the postcode that is home to both team’s stadiums, there is no clear divide in the rest of the region.
Examining the map in more detail, and looking at specific postcodes, the researchers were able to see which teams are most mentioned there. Interestingly, Oxford academics mentioned Manchester City more than any other team.
On the Oxford Internet Institute website the team of researchers commented, “There is no doubt that using Tweets as a proxy for fandom is messy and not always reliable. But, the data does give us a rough sense of who is interested in – or at least talking about – what, and where they are doing it from.” The authors added, “It allows to begin to counter myths, for example that Mancunians don’t support Manchester United, develop new insights about places that we don’t necessarily have good data about, and most importantly, have some guesses as to which team the Queen might support.”
Previously the research has focused on geo-located tweets about floods and earthquakes. They also tracked racist comments during Obama’s election in order to see what mapping the data would reveal. Whilst the researchers were measuring tweets rather than people, it was established that there are some fairly strong clustering of hate tweets centred in southeastern US states. These states had a much higher rate than the national average.
In an article for the Guardian’s Data Blog, the authors of the research concluded, “the unfortunate fact is that most states are not immune from this kind of activity. Racist behaviour, particularly directed at African Americans in the US, is all too easy to find both offline and in information space.”
One of the researchers, Dr Mark Graham commented, “This method could be used for much wider purposes: possibly, in times of crisis we could see what was being said on the social media site about hurricanes or riots to see what was actually happening ‘offline’ in certain areas of the country. This is exploratory work into the digital shadows we create, and the full potential of its applications has yet to be realised.”