Office of the Vice President for Global Communications

Tuesday, September 18, 2012

Seminars explore how learning analytics can improve student performance

Using data generated from the performance of thousands of students, U-M faculty members are exploring ways to refine their classes, and discover more about how students learn and ways to improve how they do in class.

Drawing on data generated in the course of instruction — including CTools usage information, student records, and class grades — faculty and staff have been using the new tools of learning analytics to help students be more successful.


Learning Analytics Task Force Chair Tim McKay presented resources and data currently available at U-M and previewed the task force's longer-term goals at its initial seminar. (Photo by Pam Fisher, CRLT)


More information

Schedule and registration information for the Student Learning and Analytics at Michigan series can be found online, along with presentation slides and videos.

Task force members

Tim McKay (chair), Arthur F. Thurnau Professor, professor of physics and astronomy, director of the LSA Honors Program
Steve DesJardins, professor of education
Sue Dynarski, professor of public policy and education, associate professor of economics
Barry Fishman, associate professor of education and information
Bill Gehring, Arthur F. Thurnau Professor and professor of psychology
Anne Gere, Arthur F. Thurnau Professor, Gertrude Buck Collegiate Professor of Education, professor of English language and literature
H.V. Jagadish, Bernard A. Galler Collegiate Professor of Electrical Engineering and Computer Science
Mika LaVaque-Manty, Arthur F. Thurnau Professor, associate professor of political science and philosophy
Joannna Mirecki-Millunchick, professor of materials science and engineering
Martha Pollack, vice provost for academic and budgetary affairs
Vic Strecher, Professor of health behavior and health education
Stephanie Teasley, research professor of information and director of the Usability, Support and Evaluation Lab

Earlier this year, Provost Phil Hanlon charged a Learning Analytics Task Force (LATF) with helping the university community take advantage of instructional data to better understand the relationships between student preparation, study habits, teaching approaches, and success in the classroom.

The first fruit of the task force's efforts is the Student Learning and Analytics at Michigan (SLAM) seminar series, which will meet a dozen times in 2012-13. Faculty, staff, graduate students, and postdoctoral scholars can learn how U-M faculty and visitors from other universities are analyzing data generated through students' academic activities.

Hosted by the task force, in collaboration with the Center for Research on Learning and Teaching (CRLT), the series features speakers who have built learning analytics tools to foster student success, such as Carnegie Mellon's Learning Dashboard. The next seminar will be Friday with George Siemens, one of the leading scholars in learning analytics.

At the kick-off seminar last Friday, LATF Chair Tim McKay introduced an audience of more than 50 faculty, staff and graduate students to the topic of learning analytics by surveying resources and data currently available at U-M and previewing the task force's longer-term goals.

“U-M is an ideal setting for learning analytics projects due to the wealth of data available and the expertise of a wide range of faculty,” McKay said.

Meg Noori, director of the Comprehensive Studies Program, said the recent seminar inspired several new ideas for analytics projects she would like to engage in this year, continuing work the program undertook over the summer.

Examples of U-M faculty who already are working on analytics projects include Anne Gere, director of the Sweetland Writing Center, who researched the placement system for first-year writing courses. Gere confirmed that there is alignment between incoming students’ own understanding of readiness for university-level writing and specific, quantifiable aspects of their writing samples.

McKay's own project analyzed 10 years of student performance in his introductory physics courses. His findings allowed him to develop the E2Coach system that advises students based on their unique previous academic record, student profile and current course performance. Students using E2Coach performed a quarter letter grade better than expected in a pilot study this past term.

"It's impressive to see the commitment Michigan is showing around student learning analytics — not only to advance the national conversation — but to improve outcomes for our students. There's great work already being done by our faculty, and the task force's initiatives will elevate that work to the next level," said Joe Howard, a graduate student at the School of Education.