Integrating Evaluation and Agent-Based Modeling: Rationale and an Example for Adopting Evidence-Based Practices

Abstract

Background: While there is a great deal of discussion about complex adaptive systems in the field of evaluation, little has been written about the expression of complex adaptive systems in terms of agent-based models, the execution of those models as computer simulations, or the tight integration of agent-based modeling with traditional evaluation methods. These topics need to be explored if evaluation is to move beyond using complex adaptive systems in an exclusively heuristic fashion. Purpose: This paper advocates for the integration of agent based modeling into traditional evaluation activities. We advance this position in order to spark a movement toward incorporating the formal application of complex adaptive systems into the theory and methods of evaluation. To make our case, we provide a hypothetical example of how interactions between an agent-based model and evaluation can provide unique and powerful understanding about the adoption of evidence-based practices in the field of addiction treatment. We advance our argument by addressing four questions: Why combine evaluation methods with simulation methods? Why use a complex adaptive systems approach? Why add agent-based modeling to the mix of evaluation tools? What is the relationship between agent-based modeling and the theories and methods of evaluation that are being developed to deal with surprise, unintended consequences, and program evolution? Setting: Not applicable. Intervention: Not applicable. Research Design: Not applicable. Data Collection and Analysis: Not applicable. Findings: Not applicable. Keywords: complex adaptive systems, agent-based models, evidence-based practice, unintended consequences, developmental evaluation