A Call to Action for Student Success Analytics

“Optimizing student success should be Institutional Priority #1. Effective change management should be deployed to execute this strategy and build the organizational capacity required for its success.” – Baer and Norris

Just as every change initiative requires a project plan, so determining how to optimize student success requires a plan. Leveraging analytics to optimize student success is an institutional strategy whose time has come. It has the potential to dramatically improve retention and graduation rates, reduce the total cost of completion of certificates and degrees, enhance the full spectrum development of learners, enable greater career connections and serve as a differentiator for institutions that acquire this organizational capacity.

The current environment on campuses is one of good intentions about improving student success. All campuses are held accountable for persistence and completion. However, many efforts to improve student success are fragmented and lack ongoing assessment about efficiency and effectiveness of various programs and interventions. I would ask the following questions: Do campuses know the total number of student support initiatives? Who is responsible for the initiative? What metrics are used to determine effectiveness? Is the initiative assessed? What insights are gained from review of the success of various programs? Are budgetary decisions based on insights and outcomes from program reviews? What would a comprehensive plan look like to strategically take action to improve student success analytics capacities?

Don Norris and I have been studying what makes for successful analytics capacity. There are seven dimensions that span the student life-cycle and the student experience. If viewed in total, campuses can leverage actions and interventions to substantially reduce or mitigate risk and increase student success.

Managing the Pipeline

Institutions have been utilizing descriptive/diagnostic analytics for years to shape their entering classes, refine policies, and identify at-risk students for mentoring and special interventions. Institutions are using such strategic enrollment management techniques to improve their yield/conversion rates, enhance current enrollments, and improve retention. In recent years, predictive analytics have been added to these practices. The University of Texas at Austin is recognized as a best practice leader for its use of predictive analytics to identify at-risk students and to craft mentoring and support experiences for them.

Eliminating Barriers, Obstacles and Risky Structures/Practices/Processes

Institutions have utilized descriptive and diagnostic analytics to create best practices that reduce risk in structures, processes, and policies. Favorite best practices have included reinventing the first-year experience; eliminating barriers, bottlenecks, and inconsistencies of approach; providing better mentoring and advising; and advancing peer-to-peer learning and tutoring (supplemental instruction). In addition, creating a focus on guided course and service pathways for students improves decision making over the traditional cafeteria-based choices that often overwhelm students. See Redesigning America’s Community Colleges (2015) by Bailey, Jaggers, and Jenkins for excellent overviews of the research on community colleges.

Dynamic Interventions in Real Time

For some time, proprietary institutions like the American Public University System (APUS) and Capella University have utilized embedded practice techniques; and these best practices are being deployed in many institutions’ predictive analytics foundation to identify at-risk behavior by online students. Rio Salado College developed similar alert/intervention capabilities. Civitas Learning has developed predictive analytics applications to enable institutions to provide dynamic, real-time assessment, alert and intervention capabilities to front-line faculty and staff. An important contribution of such user-friendly applications is that they enable cross- institutional teams to undertake higher-level predictive and prescriptive analytics. An important contribution of such user-friendly applications is that they enable “analytics for the masses,” which enable analytics staff to undertake higher-level predictive and prescriptive analytics. Such applications have been the foundation for the leadership of institutions like the University of Maryland University College to become recognized leaders in building cultures of evidence and student success improvement.

Integrated Planning and Advising

Over the past decade, many leading institutions have developed sophisticated student advising and pathway planning systems. Systems at Sinclair Community College, Arizona State University, among many others, have demonstrated that impressive improvements are possible when students are advised into guided, planned pathways and provided with a combination of education planning, counseling and coaching, risk targeting and intervention, and transfer and articulation guidance.

These advising systems have expanded to include dynamic, analytics-driven interventions and additional functions. The Bill & Melinda Gates Foundation has supported the development and definition of these Integrated Planning and Advising for Student Success (iPASS) work which have emerged as “an institutional capability to create shared ownership for educational progress by providing students, faculty, and staff with holistic information and services that contribute to the completion of a degree or other credential.” ¹

Civitas Learning’s Degree Map application provides advisors and students with the capabilities of seeing student progress in a major, what courses remain, or what happens when a student considers a change of major, including time and cost. The Degree Map implementation at Austin Community College led its President Richard Rhodes to say “when students have more clarity and control of their degree path, they’re significantly more likely to make progress on their learning path and, best of all, cross the finish line and complete a certificate or degree. These are the kinds of outcomes all of our student success innovations are shooting for!”

Over time, integrated systems promise to expand their functionality and become a primary multi-faceted vehicle for institutions managing the overall context of student success. For example, through the Civitas Learning predictive models, campuses can identify which interventions are working, for whom, and by what mode they should be delivered. In addition, the solutions provide insight into which inspirations can be provided to take a student to the next level.

Next Generation Personalized and Competency-Based Learning

Personalized, adaptive learning and competency-based learning promise to usher in a new era of learning analytics, which is defined by Brown, Dehoney, and Millichamp as “the measurement, collection, analysis, and reporting of data about learners and their contexts, for purposes of understanding an optimizing learning and the environments in which it occurs.” These embedded learning analytics will automatically and continuously collect data on learner progress and attainment and require far more robust analytics and management tools.

Big Data and Data Science

For a number of years, a relatively small number of institutions have been engaging in deep data dives and big data applications to illuminate issues relating to retention and student success. The introduction of data science experts enables institutions to profoundly understand student success for individuals and clusters of students to a degree never before possible. Such institutions have been able to move beyond implementing generalized best practices toward creating personalized learning experiences and interventions that optimize learning for individual students.

Data science can create predictive models that lead to prescriptive interventions. Civitas Learning is pioneering the deployment of such data science-based techniques. The results are being embedded in many of the dimensions of the student success typology. Over the next few years, these applications will support dramatic growth in “student success science,” which is bringing together data science and student success to improve information, knowledge, and action around improving student persistence and completion.

Expanding Success to Include Employability and Career

Many institutions are expanding their definition of success to include employability and career success. The Lone Star College System introduced an integrated Education and Career Positioning System (ECPS) that lets students, faculty, advisors, and parents simulate, navigate, validate, and plan a student’s education-to-career options to select the best individual journey for that student. The ECPS takes all personalized interests, values, skills, and academic records for students and distributes personalized student analytics directly to them for planning. Colorado State University’s comprehensive, holistic approach to student success considers all element of the student experience (curricular, co-curricular, and work). Over time, it is clear that IPASS systems will be expanded to manage and take account of the full range of learner developmental experiences, including entrepreneurship and innovation activities, design competitions, co-op experiences, and a myriad of others.

These dimensions provide the template for campus leadership to take a stand in improving student success in a systematic manner. Each dimension leverages the others to optimize what can be done when a campus builds a strategic call to action for student success analytics required to truly make a significant difference in student achievement, persistence and completion.

Top Banner Photo: Change and Culture by Deb Nystrom used by permission CreativeCommons BY-2.0.

—
1. Brown, M., J. Dehoney, and N. Millichamp. “What’s Next for the LMS?” EDUCAUSE Review, July/August, 2015, P.40-51. http://www.educause.edu/ero/article/whats-next-lms