Tuesday, December 16, 2008
Been talking with one of my students about doing some FaceBook analytics as a part of a research project looking at social networking. One of the things which interests me is the appropriate research methodologies to actually get to the bottom of why and how people use these sorts of networks.
Two publications in the near past (Dec 2009) give clues to why we need to get our research methods sorted.
1) Peter Schwartz in the Huffington Post blogs on the proposed death of facebook (well I paraphrase)
2) a paper from Huberman et al on Social Networks that matter: Twitter under the microscope
so my student wants to do a bit of traversing of networks and mapping patterns of connections. and it would be good if we could identify any markers for behaviours - there have been a few posts on that stuff too - something like "ten types of twitter users" I seem to recall
And after we have collected our first set of data from the network traversal, we want to talk to different user types and get some backfill on their actual usage, motivations etc etc. The whole task would be monumental - or at least suitable for some extended study, but this is going to be short and to the point.
The issue I have with both the Schwartz blog, and Huberman et al, it that they leave dangling more questions that they actually answer.
Looking at what Schwartz offers in anecdotal evidence about user preferences, what Huberman et al suggest about evidence of real network (the little networks within networks that really matter - my phrase not theirs) reminds me that I also have anecdotal and personal ethnographic evidence of a more complex set of connections, and values.
and what I want is a method of exploring this sort of network, both by physically getting the evidence from the networks, and from identifying and gathering personal data from Representative selections of those users.
I 'know' from discussions with active users whose posts I value, that there is mileage in this area, the question for me (and my student) is how are we going to get to that precious data, and therefore, what can we do to get a bit further down the methodological minefield
Now I also have an uneasy feeling that I need to be getting into a whole load of the data mining reserach.
Seems that Tesco's marketing could be very enlightening for me. Thanks to Tony Hirst @psychemedia on Twitter and OUseful on WordPress, I now have an interest in Dunnhumby and I have an extra book to take away with me for my next few days rest and recuperation :-)
And thanks to Beth Kanter I am now aware of comMetrics and their list of best 100 twitter tools!