Human Complex Systems

This paper presents a case-study demonstrating how traditional ethnographic fieldwork may benefit from analysis using computer simulation modeling tools. Paul Willis’ classic ethnography, “Learning to Labor” is represented using the system-modeling software Stella, and used experimentally to test several aspects of the worker-business relationship. I include a description of the models, the rationale for using systems modeling instead of agent-based modeling, and results from two sets of simulation experiments. I describe how the model represents ethnographic data, how it was used experimentally, and implications of the research as a critique of “Learning to Labor.” I close by suggesting that the semantic conception of theories (McKelvey 1999) can be an appropriate model of scientific investigation to link social theory, ethnographic fieldwork, and simulation modeling.