Agent-Based Modeling and Organization Studies: A critical realist perspective

摘要

The method of agent-based modeling raises philosophy of science issues that modelers have yet to resolve in a way that reconciles their work with that of other management and organization researchers. As a result, agent-based modelers have made only modest contributions to advancing organization theory and empirical research. In response, this study proposes critical realism as a philosophical perspective to understand and orient agent-based modeling research. A critical realist perspective clarifies the nature and purpose of agent-based modeling and indicates the potential complementarity between agent-based modeling and other approaches to theory building and testing in the field of management and organization studies. Key emphases within critical realism on mechanismic explanations, emergence, simplifying assumptions, and abductive reasoning support agent-based modelers' practices. Critical realism carries implications for specifying models, clarifying ontology, evaluating model outcomes, validating models, triangulating, and identifying the limits of agent-based modeling. This study provides practical guidance from a critical realist perspective-to modelers and nonmodelers-to advance the contribution of agent-based modeling to management and organization research.