New research published in Information and Learning Sciences suggests that considering the role of the teacher is critical in the design and implementation of computer-directed learning
In our new paper, "Exploring how teachers support students’ mathematical learning in computer-directed learning environments" my co-authors (Melissa Boston of Duquesne University and Mary Kay Stein from the University of Pittsburgh) and I argue for the importance of the teacher’s role in implementing cognitive tutor (CT) or other computer-directed learning environments. CT systems are often designed to take on the role of the teacher—for example, by providing students with tasks, feedback, hints, and solutions. By default, students interact with the CT system, leaving little space for the teacher. However, our research suggests that by taking on different roles in computer-directed learning environments, teachers can impact students’ engagement with the mathematics and create opportunities for students to learn. It is therefore important to consider the role of the teacher in designing and implementing computer-directed learning environments in formal education settings.
When we began work on this NSF-funded project, we had planned to focus our efforts on professional development for teachers. We intended to explore how to structure classrooms in ways that would minimize obstacles to implementing the CT system as it had been designed (e.g., with students interacting with the system with little role assigned to the teacher). What we quickly realized from our earliest observations of teachers implementing the system was that teachers seemed to be playing a critical role in the way students interacted with the CT system and the mathematics, even though the tutoring system had mostly been designed to be teacher-less. Our attention then shifted to the role of the teacher and the types of interactions they were having with students as students engaged with the cognitive tutor.
Pulling on research frameworks previously used to study the implementation of math and science curriculum and literature on cognitive tutors, our project explored the need to consider the teacher as in integral part of the implementation of cognitive tutors. The article details the iterative design and implementation of a cognitive tutoring system with 10 educators and 134 students. We were able to systematically capture and classify interactions between the student, the CT system, and teachers. This instructional triangle, based on the work of Cohen and colleagues (2003), helps us better understand that implementing a CT in formal education settings requires an understanding of the impact the educator has on system-student interactions, particularly as it influences students’ understanding of the content covered by the system (in this case, proportional reasoning).
In the article, we also describe a data collection tool we developed, the interaction tracker, which enabled us to identify four types of interaction patterns:
1. Enactment of the cognitive tutor environment as designed
2. Educator taking over the CT environment
3. Educator facilitating the CT environment
4. Educator facilitating the mathematics
Beyond defining and detailing data associated with these interaction types, our paper connects the interaction types with results of students’ assessment of proportional reasoning. These results point to clear connections between interaction patterns and types of thinking in which students were supported to engage. In light of these results, my coauthors and I discuss important next steps for this line of work.
We hope that other researchers studying the implementation of computer-directed learning environments in formal education settings consider analyzing the implementation of the system, including using formal observations of how teachers are interacting with students and the system as data. Using the interaction tracker to capture new and unique interaction types will allow for a deeper understanding of the breadth of possible interactions and how they might be leveraged or mitigated in the design of such systems to optimize students’ engagement, interest, and learning.