Model co-creation from a modeler’s perspective: Lessons learned from the collaboration between ethnographers and modelers

Abstract

This paper reports on the authors’ ongoing collaboration on model co-creation, a process that involves not only the reconciliation of methodologies (qualitative vs. quantitative), but also of epistemologies (empirical vs empirical/rationalist) and ontologies (observable referent vs. abstracted referent). The co-creation process has taken place over several months, from early 2017, both in person, teleconferencing and via email. The result was an ethnographic model of the refugee situation in Lesbos, Greece. The qualifier “ethnographic” means that the simulation’s purpose was to capture the problem situation described by ethnographers in a manner that resembles their observations, not to answer a research question. Ethnographers used the modeling process – mostly elicitation and variable identification - to think about questions they had not considered in the field. Further, the used the prototype model to further narrow their desired modeling scope and ask new questions. Lastly, notes captured by the ethnographers in the field highlight the challenges of the modeling situation.