This is a study group of the book Dynamical Processes on Complex Networks, by Alain Barrat, and a discussion group of the book's applications to Economic, Cultural and Social Anthropology as well as to other social sciences.
Latest Activity: Dec 8, 2012
This group does not have any discussions yet.
Yes, I love Douglas White's work! By the way, here is my first draft and I would greatly appreciate any feedback!Proposal.pdf
Sounds cool Johannes. That is the direction I've been advocating for a while (rather than make a network dynamic, model dynamic systems and abstract networks from the data they produce...).There have been, I recall, efforts at reproducing large-scale demographic features using local level cultural rules. See for example the work of Douglas White (http://eclectic.ss.uci.edu/~drwhite/doug.html), or this, for example, http://jasss.soc.surrey.ac.uk/1/1/1.html. No doubt, though, you are more familiar with the current literature than I am. Anyway, I know about kids. Mine want attention at the moment. Now just where are those crayons?
...Thank you John! Actually, my girl friend is of great help, so my parenting isn't purely single.
In terms of sample size, you are right. However, I'm doing something rather unusual, namely I'm hoping to use ethnography to build an Agent Based Model, which then builds a theoretical network data set (not a purely empirical one). In other words, I'm hoping to discover the rules by which the structure is built (by which society assembles itself, to use the language that computer scientists might use) and then from that, build a model that produces the structure; a simulated network data set that looks like the empirical one (one hopes), but has an arbitrary sample size. Granted, there will necessarily be some assumptions (which very traditional Anthropologists might not be comfortable with), but as you know, there is always some trade-off to be made. The assumptions can be limited (and with enough time spent in the field can be reduced) by careful observation.
Single-parenting with a 7-year old to take care of and 5 classes at Harvard and MIT. Hail, Superman. I'd be delighted to take a look at the dissertation proposal. Do send it along when it's ready.
I am struck by your comment that "I think that for purposes of gathering rich network data, no other tool is quite as powerful as ethnography." In principle, I agree. Practically speaking, I'm not so sure. Whatever else ethnography does, it sharply limits sample size. My own approach has been to start with archival data, use network analysis to demonstrate overall patterns, and use ethnography to follow-up, once I know where I want to focus.
I have also been very distracted by the fact that I was suddenly given custody of my 7 year old daughter Ayla, whom I love very much, while taking 5 classes at Harvard and MIT (some of which speak to Dynamic Networks, by the way). In terms of ethnography, I am now planning to use ethnography as a means by which to gather interesting network data (a data gathering tool). I think that for purposes of gathering rich network data, no other tool is quite as powerful as ethnography (as I don't believe that simply asking people to name their friends or asking them how they found their job is very truth revealing). Also, in my case I need to find out about how and whether people maintain their ethnic boundaries and how cross-cutting cleavages between members from various groups form (something that any other method seems inadequate for). If you don't mind, I would love to send you my dissertation proposal, once I have it in a better state. The proposed work includes the use of ethnography, agent based modeling and game theory, as well as dynamic network methods.
This may be useful also: http://intersci.ss.uci.edu/wiki/index.php/Python_for_networks
Thank you, John for your last two very interesting posts! ...I'm a bit under deadline stress these days and thus I haven't had a chance to post lately.
Johannes, thanks. And here's one for you, with a tip of the hat to Complexity Digest: http://www.bartlett.ucl.ac.uk/casa/pdf/paper170.pdf
Johannes, when you got me working on R I signed up for the R-bloggers newsletter, where today I found this new GUI for Spatial Analysis. Instantly thought of your project.
Welcome toOpen Anthropology Cooperative
Sign Upor Sign In
Or sign in with:
© 2013 Created by Keith Hart.
Report an Issue |
Terms of Service
Please check your browser settings or contact your system administrator.