Development and Validation of a Three‐Dimensional Framework for Assessing Mathematical Collaborative Problem‐Solving Ability

ABSTRACT Collaborative problem‐solving (CPS) has been one of the essential core skills for future students. However, we found that there are many studies on the evaluation of collaborative problem‐solving skills, but there is little studies on the subject areas based on which collaborative problem‐solving skills have been evaluated. The three‐dimensional evaluation model of “cognitive ability‐collaborative ability‐mathematical problem‐solving process” constructed in this paper identifies the sub‐dimensions of cognitive ability and collaborative ability and divides the sub‐dimensions into definitive levels according to the Structure of the Observed Learning Outcome (SOLO) classification theory as well as Bloom's theory of affective classification, which results in the formation of a system of observable and measurable indicators. We developed an effective structural equation modeling (SEM) approach to explore the relationship among mathematical collaborative problem‐solving（MCPS） ability elements though the utilization of Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA). The result showed that eighteen elements all serves as predictive factor for Mathematical Collaborative Problem‐Solving ability variables. Therefore, we use mathematical problem solving as a vehicle to construct a framework for assessing junior high school students’ collaborative problem solving ability in mathematics. This model provides a clear measure of MCPS skills and facilitates education and learning for both teachers and students.