Demonstrating the impact and effectiveness of educational interventions, including medium and high-fidelity simulation, has long been fraught with methodological challenges and ambiguities. This is particularly the case when there are several confounding factors and variables operating in situations where control trials are inappropriate, and investigative costs can be high. Current theoretical and empirical evidence, while emerging, is parsimonious and fails to take account of the characteristics of different modes of simulation, their contested theoretical models of learning and the opportunities presented by cutting edge computer science. Medium and high-fidelity simulations, situated within technology-rich environments, generate new forms of complex data that have the potential to provide insights into 'real-world' practices. Drawing on a range of locally based studies, we argue that until the methodological questions and data management Downloaded from systems can be addressed, the evidence to determine the judicious and optimal use of simulation to improve student and practitioner performance and patient outcomes will remain primarily reliant on proxy measures of self-efficacy and competence.