By Kurt Ewen, Assistant Vice President, Institutional Effectiveness and Planning
Much of Valencia’s innovative practices over the past 10 to 15 years have been inspired by a set of Big Ideas. One of these Big Ideas is that anyone can learn anything under the right conditions and is built on the belief that our students possess all of the biological gifts and the inherent capabilities to learn anything we teach.
When put into practice, this Big Idea shifts the focus of our work from the deficiencies of the learner to the conditions we create for learning. Over the past 10 years, online instruction has grown to nearly 20 percent of our total enrollment, yet student success in online classes is, on average, 10 percent lower than student success in face-to-face sections of similar classes.
While it may be true that online classes are not the right fit for all students, we also believe that we need to consider the conditions of learning in online courses and develop tools that might help us help students to be more successful.
Two years ago, Valencia was invited to explore how predictive analytics might be used to improve student learning and the student experience with Civitas Learning, an emerging cloud-based predictive analytics company. Civitas Learning develops tools intended to deliver personalized, real-time recommendations directly to students, faculty, advisors and administrators that inform decisions relatedtostudent learning and the possibilities of success.
Predictive analytics is commonly used by companies like Amazon to make recommendations based on large data sets associated with customer behavior online. In the context of online learning, predictive analytics might create the possibilities of an early warning system to identify struggling students, using student data readily available in our student Information System (Banner) and our Learning Management System (Blackboard).
After two years of discussion and analysis, Civitas data scientists are helping us explore student data in ways that we are unable to do using traditional analytic models. Predictive analytic models have been able to identify patterns in student behavior and engagement that are predictive of success or failure in an online course and create the opportunity for faculty to develop ways to intervene while there is still time for a student to be successful.
This summer, a group of 10 Valencia professors (all are tenured and have completed our Digital Professor Certification) agreed to participate in the predictive analytics track during Destination 2014: Innovations and Change. Faculty participants in this track agreed to the following:
- Emerge as campus-based subject matter experts in predictive analytics and how it is currently being used in learning environments.
- Directly engage data scientists and designers from Civitas Learning in order to better understand the predictive analytic tools they provide.
- Identify the potential uses of Civitas’ analytic tools to the online learning environment.
- Directly engage data on their students (enrolled in summer 2014) and work with their Destination peers to determine how best to “use” the data to improve student learning/success.
- Study and/ test the impact of student engagement strategies in online courses using predictive analytic tools to measure changes in student performance over time.
- Present their shared experience with the predictive analytic tools developed by Civitas Learning and make recommendations concerning future faculty use and development needs.
- Serve as campus-based resources in support of the scaled implementation of the Civitas predictive analytic tools in the future (if they are found to be helpful to faculty and students).
Work on this project has continued beyond Destination and will conclude at the end of the summer term. Updates on this project and possible next steps for predictive analytics at Valencia will be announced early in the fall term.