The goal of this research study has been to develop, implement, and evaluate a school reform design experiment at a continuation high school with low-income, low-performing underrepresented minority students. The complexity sciences served as a theoretical framework for this design experiment. Treating an innovative college preparatory program as a nested complex adaptive system within a larger complex adaptive system, the school, we used features of complex adaptive systems (equilibrium, emergence, self- organization, and feedback loops) as a framework to design and evaluate a strategy for school reform. We created an environment for change by pulling the school far from equilibrium using a strategy we call "purposeful perturbations" to disrupt the stable state of the school in a purposeful way. Over the four years of the study, several tipping points were reached, and we developed agent-based simulation models that capture important dynamic properties of the reform at these points. The study draws upon complexity theory in multiple ways that support improved education for low-achieving students. This research does not provide a blueprint for an education reform that can be duplicated elsewhere--instead it provides a guide to help others to navigate the turbulent waters of school reform, informed by complexity theory. The exact nature of "purposeful perturbations" will differ from setting to setting, but this technique for generating persistent change may apply more generally. This study provides an existence proof of a successful teacher-initiated educational reform and shows that complexity theory can only not just describe educational reform but can also productively guide it.