Seminar Friday, July 25, 3:30-5p, Michael Agar, "Bugs and Fieldwork: Ethnography and Agent-Based Modeling", FriamGroup Applied Complexity Lecture

[Friam mailing list]

Parking will be limited at SFI this date. Please try to cluster into
carpools.

Thanks for the afternoon slot, Jose!

** FRIAM Group Applied Complexity lecture: Friday July 25, 3:30-5:00 p.m. **

Location: Medium Conference Room

Topic: Bugs and Fieldwork: Ethnography and Agent-Based Modeling

Speaker: Michael Agar

Affiliation: Senior Research Scientist, Friends Social Research Center
        Professor Emeritus, University of Maryland, College Park

*** Abstract: ***

Ethnographic research, with its century of history in anthropology and
sociology, is the only social research that assumes nonlinear dynamic
systems, both as research process and research product. Artificial
societies, a version of agent-based modeling, simplify and represent just
those kinds of social worlds. What sort of conversation can this old mode of
research and this new mode of modeling have with each other?

First ethnography will be described, on the assumption that most will have
had little experience with it. The argument will be that, if you were going
to look at a situation with ABM goals in mind, ethnography is the way you’d
do it. Next, an artificial society that demonstrates the strengths and
pitfalls of tacking back and forth between ethnographic research and an ABM
will be presented. (An earlier version is available in Complexity—“Drugmart:
Heroin Epidemics as Complex Adaptive Systems,” 7 (5):44-52, 2002.) This
model results from an NIH funded project to explain illicit drug epidemics.
The model tries to explain epidemiologic incidence curves of illicit drug
use as an emergent property of autonomous agent experiences and the stories
they tell as a result. The argument will be that the ethnography/model link
works, sort of. The “sort of” is the most interesting part and it will be
explored. Finally, some ways that a conversation between ethnography and
artificial societies can be mutually beneficial will be explored. For
example, artificial societies need better validation. As another example,
ethnographies need to solve the problem of the limited case study.