By Shilo Rea / 412-268-6094 / firstname.lastname@example.org
When Bill Buttlar, a professor and associate dean in the University of Illinois at Urbana-Champaign's (UIUC) College of Engineering, and his colleagues were charged with revamping their graduate program, they decided to attend Carnegie Mellon University's LearnLab Summer School.
Their reason was simple: They wanted to learn how to effectively implement cognitive tutors into their coursework, and CMU researchers have pioneered this work, a hallmark of the university's Simon Initiative.
LearnLab, a Science of Learning Center funded by the National Science Foundation, holds the intensive one-week course to provide attendees with hands-on experience in designing, setting up and running technology-enhanced learning (TEL) experiments — as well as analyzing data from those experiments. More than 50 professionals and graduate students from around the world participated in the 11th annual Summer School, which had three tracks centered on intelligent tutor development, computer supported collaborative learning and educational data mining.
For their projects, the attendees from UIUC used Cognitive Tutor Authoring Tools (CTAT) to create tutor-assisted, online problem sets for courses in the professional master of engineering programs. They also worked to determine the potential use of CTAT to track student progress as they move through tutor-based problem sets.
"Both projects demonstrated that CTAT would be an effective tool to help us scale up our online and on-campus professional engineering programs," Buttlar said. "I believe our faculty will be enthusiastic about implementing this cutting-edge technology in their courses, and moreover, I believe our students will greatly benefit from having access to problem sets guided by the CTAT intelligent tutor capabilities."
LearnLab Director Ken Koedinger thought this year's session was the best yet.
"I was really impressed overall with the projects and the questions that were asked," said Koedinger, professor of human-computer interaction and psychology at CMU. "I think it reflects how our learning science and engineering research has matured. Our tools and techniques have developed so people can really hit the ground running and accomplish a lot in one week."
For example, Koedinger said that students within the educational data-mining track no longer have to wish for particular data sets. LearnLab's DataShop is a repository of educational technology data that is so massive now that students find existing data of interest. Or, if they have their own data, DataShop makes it easy to upload that data and then take advantage of built-in analytics, making student projects directly applicable to their own interests.
Projects ranged from detecting students' misconceptions in an online physics game to whether general or targeted feedback makes larger impacts on student learning outcomes.
Nick Diana, an incoming CMU human-computer interaction graduate student, worked on generating a model that could predict if a student is likely to pass or fail a course early in the semester.
"The hope is that the model could be used to identify students on an undesirable trajectory, and intervene soon enough that they could recover," Diana said.
Ningyu Zhang, a doctoral student in computer science at Vanderbilt University, was thrilled with his experience. He attended the Summer School to learn more about intelligent learning environments, specifically how to choose the theoretical framework, code the software and then assess its effectiveness.
"I feel that going there only once isn't enough for me," he said. "There is so much that I can take away."
Carnegie Mellon researchers have spent decades advancing learning engineering in order to improve outcomes for their students. They have also taken that work and created tools and methods so that any person or institution can adopt them. The Simon Initiative, named for the late CMU Nobel Laureate professor and co-founder of artificial intelligence Herbert Simon, harnesses this cross-disciplinary ecosystem of learning science at CMU, with the goal of measurably improving student learning outcomes.