The concept of collaborative problem solving (CPS), i.e. two or more agents working to pursue a jointly valued solution, is a foundational element of social and workplace interaction. While it was certainly possible, indeed common for collaborative problem solving to occur over great distance and time in the past, the advent of the Internet and the proliferation of reliable software for remote, real-time collaboration have dramatically changed the scale and scope of what it means to work together. Individuals and industries will face significant changes, but also tremendous opportunities as they endeavor to meet the challenges present in our hyper-connected 21st century. A natural ramification of this evolution is the desire amongst individuals, educators, and employers to have an accessible and reliable measure of the skills and attributes necessary for achieving success in a deeply and increasingly interconnected society.
Developing an accurate assessment of a CPS task requires identifying, measuring and inferring skills and attributes, both on an individual and group basis, critical for successful CPS. Moreover, the type and degree of interaction the group expresses during the CPS task has to be evaluated and deconstructed in order to disentangle the individual’s contributions from the group’s contributions. Although researchers have benefitted greatly from advances in software and hardware allowing them to examine CPS tasks in unprecedented detail, they are faced with the daunting task of analyzing the truly massive amounts of information available to them. In order to make sense of the collected data, researchers are integrating research and tools from areas like computer science, stochastic theory, and theory driven psychometrics into the emerging field of Computational Psychometrics. Several recent publications co-edited by ACTNext’s vice president, Dr. Alina von Davier, take stock of the current state of the science in effort to focus the discussion: