Comparing Facebook networks with the students
On Tuesday the students in my upper-level undergraduate course on social media presented their visualisations of their Facebook networks, and it was so interesting! There is quite a broad variety in how these networks look. Some students had large but very dense networks where everyone knew everyone, and had some trouble finding clusters in the network to actually be able to see anything. Others had very clearly separated networks. Mine is one of those, although there are certainly a lot of interconnections. That big pink spot is Scott, of course.
One of the most interesting things, I thought, was looking at the four or five students who either were currently on exchange here or had been on exchange abroad. Their Facebook networks were clearly influenced by this, with very distinct clusters of friends from the exchange, but these networks were very different. The French and German students currently here had large groups of new friends made here, and one had colourcoded these by interfact language for Facebook, which very clearly shows the great cultural diversity in this group. Students from here who had been in the US on exchange had had quite different experiences, at least as reflected by their Facebook networks – which of course do not tell the whole truth. A student who went to Berkeley had a huge group of Norwegian friends from her studies at Berkeley, and beside them, in a separate cluster, a small group of Americans who had been her neighbours rather than her fellow students. I guess a lot of Norwegians go on exchange to Berkeley. A student who went to New York went to a small school and only had a few fellow students on Facebook. They both said they hadn’t used Facebook much in meeting new, American friends, although they were on exchange within the last couple of years. The European students coming to Bergen said (and their networks displayed it) that in contrast to that, Facebook is hugely important socially for exchange students here.
I’m not going to show you the students’ networks yet because those networks belong to them (although I might link to one…). But I got so excited about seeing them all together and seeing the kinds of comparisons we could make that at least some of us are going to write a collaborative paper on this. I’ve already submitted the notification form to NSD, our ethical review board, so we will have to wait until I hear back from them before we continue work on this. Here’s how I filled out the form – first time I’ve done this, and probably not the most tidy way of doing it, but there you go.
I think it could be a really interesting experiment not just in the analysis and comparison of different peoples’ Facebook networks but also in a mode of student-active research that’s quite appealing. Obviously you are the people best qualified to analyse your own networks – you’re the only person who knows all the people in it and how you know them. So if we get a bunch of smart people who are reading lots of social media theory and analyse our own networks and compare them, that sounds pretty interesting, don’t you think? We’ll see where it takes us.
We used Netvizz to download our networks and Gephi to analyse them. Both these tools are brilliant. Netvizz is developed by Bernhard Rieder at the Digital Methods Initiative at the University of Amsterdam, and is simply a Facebook app (so search for it on Facebook) that lets you download data about your network or pages and groups you are on. You can open the data up in Gephi, which is an open source cross-platform piece of software for social network analysis. I found Gephi a little hard to get started in (hint: you zoom on a mac by moving two fingers up and down on the touchpad and if you lose your window find it again in the window menu at the top of the screen) but really easy and powerful once you’re past getting annoyed at the interface. The students were able to independently make and analyse their Facebook networks after a three hours workshop playing with the tools, so it works well in the classroom.