Dive Brief:
- A report from New America reveals the potential gains and losses from institutions using predictive analytics as a support tool for low-income students. The Promise and Peril of Predictive Analytics in Higher Education: A Landscape Analysis examines the possibility of discriminatory approaches to students' degree and career goals, based upon analytical data.
- Several institutions, including Temple University, Georgia State University and Austin Peay State University were highlighted for programs which highlighted students on the verge of withdrawal or failure, which allowed for immediate and intrusive action from academic advisors, but certain statistical analysis shows that race and class almost always yield unfavorable displays of minority student performance, without accounting for external factors in assessment.
- The risk of institutions using data to make recommendations about degree pursuits, without telling students that the programs are used for such assessments, is a breach of transparency and academic trust.
Dive Insight:
While many institutions are scrambling to discover ways to aid low-income and minority students, there is a fine line between support with bias and support with cognitive awareness of real student challenges. Data can reveal that a student is suffering in a particular class or major, which could lead to conversations about transferring or dropping out.
But the data doesn't reveal information about hunger, or the need for certain students to work while enrolled full-time, which can impact performance. It is up to college leaders to contextualize analytical tools, and to ensure that advisors represent a diverse set of perspectives and knowledge base relative to student success.