Predictive Analytics in Higher Education: Five Guiding Practices for Ethical Use

Organization: New America
Author: Manuela Ekowo and Iris Palmer
Published: March 1, 2017

This report discusses the use of data in higher education, which are primarily used in the following ways:

• Early-Alert Systems. In an early-alert system, flags are triggered based on academic and nonacademic data from students that signal when they may need additional support. Academic interventions may include tutoring, meetings with an adviser, or assigning a coach or mentor to the student. For non-academic flags, colleges can deploy financial supports (i.e., emergency grants) or referrals to other supports (i.e., mental and medical health, child care, transportation, housing, and food).

• Recommender Systems. Recommender systems allow students to plan or map their degree, and integrate transfer credits or prior learning assessments into that plan. One common use for recommender systems is helping students choose courses to take next and/or choose a major based on data about their previous academic performance.

• Adaptive Technologies. Adaptive tools use data on how students learn to customize the learning environment for each individual student by identifying gaps in knowledge, skills, and abilities and adjusting content delivery to support deeper and more efficient learning. • Enrollment Management. Enrollment managers use algorithms (computer-based rules) to decide how best to target recruitment efforts and distribute financial aid.