News - Using Predictive Analytics for Student Success

Using Predictive Analytics for Student Success

Using Predictive Analytics for Student Success

By Rhonda Gregory, originally posted on the Desire2Learn blog.

To cultivate an environment where learners can thrive, institutions require innovative technology that will help create the conditions needed for them to succeed. The challenge is deciding how to start. How can an institution develop practical strategies to identify at-risk learners and help them stay on a successful path to graduation?

Focused on a way to use historical and current learning data to help identify which students are potentially at risk for dropping out, withdrawing, or failing altogether, Greenville College is using predictive analytics technology to dive deep into key achievement, engagement and completion data to get a new perspective of learners’ progress and success. Greenville College wanted a way to use existing data to help identify those students who might be struggling, much earlier in the year and more effectively than traditional methods had provided.

As a Christian liberal arts college, Greenville College is committed to transforming lives and helping students maintain success in their courses, stay in their program, and complete their degree. So, in the Fall of 2013, the school began using the Desire2Learn Insights™ and Student Success System™ solutions to tap into its learning environment data and understand what was possible for predicting students’ success. The outcomes of this venture are generating a buzz. Thanks in part to Greenville College’s implementation of the Student Success System, the tool is a finalist for the Learning Impact Award provided by IMS Global Learning Consortium recognizing outstanding applications of technology that address the most significant challenges facing global education and learning industries.

Greenville College began its validation by targeting primarily high-enrollment, freshmen courses in the school’s Desire2Learn Learning Environment. The college used the student grades and engagement information from the current courses where predictive modeling had been turned on. After some initial tweaking and adjustments, the findings were remarkable — not just regarding the impact of the predictions on student success but also yielding valuable insights gained about existing procedures for at-risk interventions as well as learning environment usage patterns.

In very short order, Greenville College was able to demonstrate the advantages of the Student Success System to faculty and begin developing a best practices protocol for using and engaging with the learning environment. Faculty immediately saw value in the success trending indicators (up and down arrows) most notably, in its at-a-glance ability to direct them to students declining in success. For those instructors with fifty or more students in their class, a tool that allows them to see students at risk sooner than they could normally identify was particularly exciting!

While faculty feedback of the Student Success System has been very positive—straightforward and user-friendly were common themes—developers took the fall semester to train staff on the proper use of the tool and to increase understanding of the predictive modeling it provides. Because the data in this particular tool relies heavily on learning environment usage, the predictions are more accurate when both learners and instructors have used the environment consistently over multiple semesters. To that end, with the ease of implementation of the predictive tool, the 2014 focus is on rolling out the Student Success System to more courses under more instructors to continually refine what student success means across all programs at Greenville College. Additionally, the school is working on a beta implementation of a Student Success System component that taps into the data from its student information system—such as students’ GPA, ACT/SAT scores, etc.—to further tune or refine each individual student’s success index.

While it is still in the early days for Greenville College with this pioneering technology, the promise of predictive analytics to deliver earlier indicators of academic struggle than traditional methods is driving new opportunities for faculty to understand the engagement, performance, and achievement of learners. By recognizing students’ potential for success or failure early in a course, instructors, advisors, and other college support staff at Greenville Collage are given the precious gift of time—time to reach out and help struggling students find their way and improve their chances for success.

Read more about technology for teaching at or

About Rhonda Gregory

Rhonda Gregory is the Director of Instructional Technology at Greenville College. In this role, she oversees the training and support of faculty, staff, and students in the use of technology for teaching and learning. She also acts as a technical consultant for online instructional design and on-ground classroom technology. Rhonda has been supporting students and faculty at Greenville College since 2006 and is also a licensed secondary teacher. She teaches college courses in education, educational technology, and literature. When not online, Rhonda enjoys spending time with family, scrapbooking, reading, and helping out at church events.

This story was published on March 11, 2014

comments powered by Disqus

Related News

315 E. College Avenue
Greenville, Illinois 62246-1145
800.345.4440 / 618.664.7100
Contact Us © 2000-18 Greenville University. All rights reserved.
Greenville University is authorized to operate as a postsecondary educational institution by the Illinois Board of Higher Education
Accreditation | IBHE Complaint Form