Scaling Instructor-driven Personal Support Actions
Year: 2015
Type: article
Abstract: In conventional learning environments, instructors are typically faced with the trade-off between the quality of the student experience and the workload in terms of hours dedicated to deliver that experience.This trade-off is exacerbated in the context of courses with large student cohorts.In these scenarios, instructors are fully aware of strategies that would increase the quality of a course such as frequent, personalised feedback, but the required resources significantly reduce (if not completely remove) the catalog of support actions.In the ideal one-to-one scenario instructors establish a conversation with the student to gain insight about their current state in terms of the learning objectives, select the most appropriate support action, and deliver it to the student usually verbally, visually or both.If we try to scale this scenario, all three stages face significant challenges if we aim at maintaining a completely human-based approach.Technology may provide support when scaling these three steps: gaining insight, selecting a support action, and delivering it to the student.But with it comes to technology, a second trade-off appears, that of the effectiveness of a tool versus its complexity of creation and deployment.A tool may provide an effective scaling of one of these stages, but they are very complex to design and/or equally complex to gain adoption by the regular instructor.The step that has received the highest amount of attention in the research community is the second: choosing the most appropriate support actions.The areas of intelligent tutoring systems and artificial intelligence for education contain multiple approaches to automate this decision based on representations of the state of the student, the domain of knowledge, the available resources, etc.
Authors: Jürgen Schulte, Andrés Pardo
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
SDG: Quality education
Open Access status: green
OpenAlex ID: W2586084404
