A methodology for assessing learning in complex and ill‐structured task domains

Abstract

New information and communications technologies and research in cognitive science have led to new ways to think about and implement learning environments. Among these new approaches to instruction and new methods to support learning and performance is an interest in and emphasis on complex subject matter (e.g., complex and dynamic systems involving things such as crisis management, environmental planning, social policy formulation, etc.). Consistent with the notion that learning involves observable changes in abilities, attitudes, beliefs, knowledge, mental models, and skills is the requirement of development methods to assess progress of learning in these complex and ill‐structured task domains. A central problem for assessment in complex task domains is determining performance standards on representative tasks against which to measure progress since ordinary problems with single solutions and approved solution approaches are generally not useful. A framework for assessing learning and performance that addresses this issue is presented along with findings from its application in three complex task domains: biology, engineering design and medical diagnosis.