Understanding student teachers? collaborative problem solving competency: Insights from process data and multidimensional item response theory

Collaborative problem solving competency is one of the core competencies that digital citizens should possess in the 21st century. There has been a growing research interest in examining learners' collaborative problem solving competency in collaborative learning tasks. However, there are still limited research studies on the composition of collaborative problem solving competency, especially considering the interactive, interdependent, and periodic characteristics of collaboration. By using the procedural data generated in collaboration, the multidimensional item response theory and the hierarchical linear model, the present study examined the char-acteristics of student teachers' collaborative problem solving competency in two real online learning tasks in a sample of 42 student teachers. The results suggested that the bigram-indicator may be more representative of student teachers' collaborative problem solving competency than the unigram-indicator. Moreover, student teachers' collaborative problem solving competency in the two online learning tasks was a five-factor structure. In addition, the social and cognitive dimensions of collaborative problem solving competency were unfolded orderly in collaborative tasks. Finally, the conclusions, implications, limitations, and future research directions were discussed.