Predicting Student Success: Beyond the Traditional Approach

illustration of an article

Given increasing competition, shifts in student enrollment, and reduced resource levels, it’s critical that colleges and universities recruit and retain the students who are most likely to succeed at their institutions. By transitioning from a risk-based model for predicting student enrollment and retention to a success-based model, you can look across the student life cycle to identify not only the factors that impede desired outcomes such as yield and student retention, but also the positive factors that contribute to those outcomes. What are the shared characteristics of students who enroll and persist? What were the shared characteristics and behaviors of your most engaged alumni when they were students? Success leaves clues.

Identifying these indicators of success can help inform smarter investments in recruitment and intervention, and can inform cross-campus efforts to maximize yield, student success, and alumni engagement.

Exploring this approach further, Rob Durkle, the University of Dayton's assistant vice president of enrollment management, and Bernadette Jungblut, West Virginia University's director of assessment and retention, recently joined Academic Impressions for a complimentary webcast offering examples of how specific indicators of success might drive specific decisions on campus.

Here is One Example: "Grit"

During the webcast, Bernadette Jungblut directed attention to the growing body of research confirming that a student's resilience -- or, more colloquially, "grit" -- is often predictive of their academic performance and persistence. In fact, even if an entering student does not look "likely to succeed" on paper (according to more traditional predicitve indicators such as standardized test scores and high school GPA), if the student scores high enough on a survey instrument designed to assess their "grit," they may actually be likelier to succeed than other students who demonstrate a strong high school GPA but are less prepared for the workload and the stress of pursuing a college degree.

Watch the complimentary webcast for a few examples of questions that you might ask on a quick survey.


For more context on resilience and other non-cognitive indicators of academic success and persistence, read our previous article: "Predicting Student Success: When SAT and GPA Are Not Enough."   This article interviews one of the foremost thinkers on the subject and includes a list of third-party instruments for testing non-cognitive indicators.

Examples of home-grown student readiness scales have been developed by W. Kent Barnds at Augustana College and by Angela Duckworth at the University of Pennsylvania.

Suppose that you either craft a home-grown instrument or adopt a third-party instrument for assessing resilience and other "non-cognitive" indicators. How might the data you collect inform particular decisions?

Jungblut offers this scenario: Imagine that you find that you have many "non-resilient" students enrolling this year. You can alert your advisors to offer more proactive outreach to the incoming class -- or to a certain cohort. You could provide data on resilience to academic success coaches, or use the data to take a more targeted approach to directing admits to a bridge program before their first term begins.