Personalized intelligent software responses for engineering students

Abstract

We developed and evaluated tutoring software called CIRCE that supports students through introductory circuit analysis laboratories and provides immediate feedback and adaptive support. The software addresses many issues with laboratory-based learning, such as students receiving feedback a week or more after the lab. Tutoring help is available at “teachable moments” during the lab and intelligent models of the student support hands-on learning and keep students on task with immediate feedback, real-time error correction and final student grades. Students perform their usual hands-on experiments using physical circuitry while submitting data to the computer in real time. For example, students sketch oscilloscope waves into the software, plot graphs or record data in tables, receiving feedback about whether their data is within or outside the expected range. We provide results of the software and its effect on student learning and affect. Quantitative and qualitative studies show that the software leads to improved learning, verbalization and conceptual knowledge; it also relieves teachers' workload. Results showed improved student ability as measured by lab grades.