From enrollment forecasts to attrition alerts, predictive models offer exciting opportunities for improved decision-making in higher education. However, modeling requires careful planning and design, especially as it relates to fostering equitable outcomes.
Institutions must ensure that predictive models are designed and implemented with equity in mind. This post will review four key characteristics of equitable predictive models. While these guidelines do not cover every concern your institution might have regarding equity in predictive modeling, they serve as a helpful gut check when reviewing existing models or building new ones.
8 Essential Predictive Models for Higher Ed
Explore the role of predictive modeling across the student lifecycle
What is predictive modeling?
Predictive modeling makes predictions about probable future outcomes based on trends and patterns in past data. A model’s predictions can inform big-picture decisions such as annual retention strategies, as well as day-to-day decisions like prioritizing outreach to individual enrollment prospects.
4 characteristics of equitable predictive models for higher education
As a Strategic Writer, Earl highlights the power of integrated higher ed data and student success technology. His work explores the benefits of data governance, predictive modeling, and data democratization across campus. At EAB and in previous roles, Earl has partnered with hundreds of higher ed professionals to develop case studies, webinars, and other content. Through this work, he gained a deep appreciation for the challenges higher ed professionals face in working with data, whether in enrollment, student success, institutional research, advancement, or elsewhere on campus. In his spare time, Earl enjoys hiking in the mountains of the Northeast and spending time on the seacoast of Maine. He also enjoys reading, writing short fiction, and sampling New England’s many excellent restaurants.