Modeling for enhanced understanding of complex systems with policy-oriented implications sometimes requires that several different levels of aggregation be considered and formally included. In the system dynamics tradition, different levels of aggregation are not normally combined, leaving certain classes of problems outside of the traditional use of system dynamics models. Agent-based models can capture a very fine-grained level of detail of the system under study but lack the ability to parsimoniously and clearly link behavior to structure. This paper presents a domain in which a combined approach seems to be adequate. Additionally, two alternatives to dealing with the problem of integrating data from different levels of aggregation using system dynamics and agent-based models are discussed.