RESUMO
Traditional modes of evaluation can be constraining when seeking to improve persistent societal problems within complex systems. Frameworks that assume stability and predictability may not match the system dynamics within which a program resides. This paper seeks to further evaluators' understanding of how to improve problems in complex systems by offering a "problem-bound" evaluation approach. This model is grounded in theoretical concepts from complexity science, complex adaptive systems, systems thinking, and improvement science. As such, it provides an agile and flexible framework for continually learning from emergent findings related to the problem. By offering this model, we seek to contribute to the limited knowledge base on how to apply systems change theoretical ideas in evaluation practice, along with practical implications regarding the evaluator's role.
