Analysing networks

The Berliner Morgenpost (Julius Troeger) presents a visually beautiful example of network analysis. It’s the twitter traffic of 61 members of the Berlin parliament with  people outside the parliament .

The 61 members were in contact with 3000 users (blue points) and sent 60000 tweets. Persons (blue points) in the middle of the graph had the most intensive contact with the members of parliament (orange, green, black according to party membership).

It’s an interesting example of datajournalism. But it’s far from giving qualitative insights: Which topics are hot? Which impacts has this use of short-message communication on forming opinions and decisions? A huge field!

An open source tool for network analysis is gephi, ‘a tool for people that have to explore and understand graphs …. a complementary tool to traditional statistics, as visual thinking with interactive interfaces is now recognized to facilitate reasoning‘.

It’s not often (if ever) used in official statistics where network analysis is not very frequent (true?).