On April 11-12, thirty-six people from nine different Penn State campuses virtually attended via Adobe Connect the EDUCAUSE Learning Initiative (ELI) Online Spring Focus Session on learning analytics. The session’s timing was perfect, as learning analytics is something that Penn State is going to hear a lot more about in the future, according to Bart Pursel, undergraduate education and instructional research associate with the Schreyer Institute for Teaching Excellence.

“The ELI event on learning analytics was perfect in terms of timing and content,” Pursel said, “As more people around PSU hear the phrase ‘learning analytics’, this ELI event provided both a common understanding of it, as well as provided a wide variety of examples through the various sessions.”

Learning analytics is defined by ELI as “the collection and analysis of data associated with student learning. The analysis of this data, coming from a variety of sources like the LMS, the library, and the student information system, helps us observe and understand learning behaviors in order to enable appropriate interventions.” Chris Millet, assistant director with Education Technology Services (ETS), said that this kind of information is beneficial to both faculty and students.

“Administrators can see how a program is doing and what areas are in need of enhancement.” —Chris Millet

“Learning analytics gives everyone a lot of data to use, and faculty can use this data for identifying those students who need an academic intervention,” Millet said. “The data in turn can be used to get a better idea of how they are doing in a course or overall in a particular semester.”

Millet added that learning analytics also has benefits for administrators. “Administrators can see how a program is doing and what areas are in need of enhancement,” he said. “This can also help with student retention since you can identify problem areas and help students achieve.”

Overall, Pursel said, learning analytics has many potential benefits. “The biggest benefit is the ability to identify at-risk students as early as possible. In some courses that only have three exams, a student might not realize he/she is in trouble until the midpoint of the semester,” he said. “With learning analytics, you might be able to pinpoint at-risk students much earlier, and therefore intervene earlier.”

“I believe some tools might also help faculty identify areas of a course that might be challenging for students,” Pursel said. “Thus, the data might influence faculty to revise certain parts of a course that students are finding confusing or disorienting.”

Pursel said that Penn State is still in the early stages of work on learning analytics. “I think it’s too early to tell what approach Penn State will take towards learning analytics. We do already have one learning analytics system in place, the Early Progress Reporting (EPR) tool,” he said. “This isn’t as automated or complex as many of the other examples of learning analytics that we heard about via the ELI sessions, but it does do the same types of things (early risk assessments for students) that many other tools do.”

To develop a learning analytics program, a variety of experts need to come together, Pursel said. Most obvious is faculty, and this includes faculty training, since they will need to understand learning analytic tools and how best to use them. Staff can help as well.

“Learning analytics can be very powerful and our statistical models can explain trends and patterns in the majority of students, but each student is unique.” —Bart Pursel

“Depending on what type of data you want to include in your predictive models, you might include library personnel, student affairs, undergraduate education, learning services and so on,” Pursel said. “All these units have different data that might be useful to better understand how a student is progressing.”

Pursel did emphasize that learning analytics is not a magic bullet to solve all academic issues a student may have. “Learning analytics can be very powerful and our statistical models can explain trends and patterns in the majority of students, but each student is unique,” he said. “The data coming out of a learning analytics system needs to be used in conjunction with person-to-person contact, as data by itself can be misleading.”

Pursel notes that there are only small groups around campus beginning to discuss learning analytics, but efforts to explore it are just beginning. “We hope that by the end of the summer, we can build a community of Penn State faculty interested in helping explore LA tools, and guide the university in implementing a tool that is the most helpful to faculty,” he said.


Tagged with →  
Share →
Skip to toolbar