Analyzing student behavior in inquiry-based learning activities using interactive simulations
Interactive simulations are increasingly being used to foster inquiry-based learning at different levels of education. However, the open and unstructured nature of such environments may often appear overwhelming, eventually leading to suboptimal learning experiences, in which for instance, students apply uninformed trial-and-error strategies to approach the tasks.
Adaptive support represents a way to provide students with better guidance in such environments, based on computational models of their cognitive processes. However, building models that appropriately capture the complexity of the processes in these interactive environments can be challenging. In this context, more research is needed to understand which behaviors are conducive to learning.
The objective of this project is to analyze the user interaction data of undergraduate physics students experimenting with the PhET capacitor lab, an interactive simulation developed at the University of Colorado Boulder. The students used the simulation to study the mathematical relationships between the different elements of a plate capacitor, such as plate area, plate separation, stored energy etc.
By applying educational data mining methods, this project aims at determining specific behavior patterns that are more likely to emerge in successful learning experiences. On the other hand, the analyses may also allow to identify possible “at-risk” students, helping teachers to plan possible interventions in a timely and efficient manner. In a further step, these findings can then also be leveraged to implement systems that provide adaptive support in an autonomous way, to further optimize students’ independent learning.