Modeling the relationship between students' prior knowledge, causal reasoning processes, and quality of causal maps

Abstract

Despite the 30 years of research on the effects of using computer-based mapping tools to construct concept maps, the research has provided little insight into the cognitive processes and strategies students use to create more accurate maps and the types of behaviors that can be mined in computer-based mapping tools to reveal and support processes that improve student performance. This study used video analysis to identify, operationally define, and code processes of reasoning students used while using a computer-based tool to construct causal maps. The process data was then used to conduct a path analysis to determine which and how specific processes impact the quality of students' causal maps and to determine how students’ prior knowledge on the given problems influence which processes students use. The findings reveal that students exhibiting the tendency to use more backward than forward processing and use more breadth than depth-first processing produced higher quality causal maps, and that the influence of prior content knowledge on what processes students use are not conclusive. The findings and the methods developed in this study present new ideas on ways to integrate learning analytics into computer-based mapping tools to model and manage the cognitive processes students use to construct better cognitive maps.