This paper describes a simulation environment that can be used to integrate population-level dynamics with those occurring at an individual, or agent-based, level. The benefit of this approach is that individual agent behaviour may be mapped at a detailed level, using differential equations, and aggregated over the entire population in order to determine population-level dynamics. Furthermore, individual agents can interact with one another, in terms of a social network structure. The environment is firmly grounded in the system dynamics approach, and, unlike conventional agent-based simulation environments, programming is not required in order to specify agent interactions and behaviours. The approach is validated by using a case study based on market dynamics. The overall benefits of the approach are summarised, and future work discussed.