Helping students acquire better self-regulated learning skills bares the promise of promoting future independent learning. However, current educational technologies fall short of interpreting, assessing, and supporting students’ use of self-regulation strategies. While students’ actions in problem-solving environments can reveal their domain knowledge, inferring their strategy use remains a challenging task. In this talk I will discuss how learning analytics and educational data mining can complement other methodologies to design and evaluate support for students’ help-seeking skills.

First, I will describe an iterative process to develop an automated, unobtrusive assessment of students’ help-seeking behaviors in the context of problem-solving environments. Second, I will describe a series of studies that showed that adaptive feedback triggered by the automated assessment helped students to acquire better help-seeking skills in a manner that transfers to novel, unsupported, learning activities. Last, I will reflect on the use of educational data mining and learning analytics to support research in the Pasteur Quadrant (Stokes, 1997), that is, help students to acquire better metacognitive skills and improve our understanding of the nature of learning.

Ido Roll is a research associate in the Carl Wieman Science Education Initiative, the Department of Physics and Astronomy, and the Department of Educational and Counselling Psychology and special education in the University of British Columbia. Ido graduated from the Human-Computer Interaction Institute and the Program for Interdisciplinary Education Research in Carnegie Mellon University.

Ido’s research focuses on helping students to become more capable, curious, creative, and collaborative learners, using interactive learning environments. He is particularly interested in using fine-grain data to understand and promote self-regulation and scientific-inquiry skills in the context of authentic environments and tasks. Ido has published numerous papers in the fields of science education, the learning sciences, cognitive science, artificial intelligence, learning analytics, educational data mining, and human-computer interaction. His work has received numerous awards in peer-reviewed conferences.