Data-Driven Educational Improvement: Modelling Learners to Support Decision-Making

Carrie Demmans Epp - Postdoctoral Research Associate, Learning Research and Development Center, University of Pittsburgh

March 16, 2017, 11 a.m. - March 16, 2017, 12:30 p.m.

McConnel Engineering room 103


Recent technological advances present a unique opportunity to support student learning through the modelling of learners, the subsequent data-driven adaptation of learning experiences, and the communication of these models. While technologies that rely on this class of approaches are being increasingly employed, their widespread use has yet to be realized due to a variety of technical, social, and logistical barriers. These barriers include model imprecision or inaccuracy, users’ limited data-literacy, and a lack of user trust in the system. I will discuss my efforts towards increasing our understanding of how these technologies fit into existing educational praxis so that we can improve both the technology and its use. More specifically, I will highlight a recent project that explores how we can support the comprehension of lay users by communicating the limitations of analytics and models. The methods employed hold the potential to increase user trust in adaptive systems and enable users to make informed decisions about how to best support learning. This type of data-driven decision-making will allow students and instructors to monitor and adjust learning in a timely manner to ensure positive learning outcomes and experiences.



Carrie Demmans Epp is a postdoctoral research associate at the Learning Research and Development Center of the University of Pittsburgh. Before moving to Pittsburgh, Carrie held Weston and Walter C. Sumner Memorial Fellowships. She was also a visiting researcher with the Open Learner Models at Birmingham group (UK) and the Graduate School of Language, Communication, and Culture at Kwansei Gakuin University in Japan. She earned her PhD from the University of Toronto, where she developed an adaptive mobile-assisted language-learning tool and explored its use. While earning her MSc at the University of Saskatchewan, Carrie integrated her undergraduate studies in Russian and Computer Science by building an adaptive pronunciation tutor that supported student motivation.

Carrie’s work focuses on the development and study of adaptive educational technologies and the mechanisms that are used to provide feedback to learners within these environments. Her work integrates human-computer interaction, artificial intelligence, psychology, and education to support a variety of populations that include university students, underprivileged children, students in special education settings, and language learners.