Identifying students at risk of dropping out early
Universities notice disengagement only after failed exams, when interventions are less effective.
5 upvotes
Problem
LMS activity, attendance, and assessment signals are siloed; advisors lack ranked outreach lists.
Current workflow
Advisors wait for student-initiated meetings or end-of-term grade reviews.
Consequences
Higher dropout and wasted student debt.
How AI might help
Risk models can prioritize outreach with explanations advisors can act on.
Limits & risks
Proxy variables may reflect socioeconomic bias.
Stigmatization; discriminatory targeting.
Alternatives today
Mandatory advising and early-alert rules.
Rules are coarse; advising capacity limited without prioritization.
Evidence
- UNESCO higher education equity briefs
Links early support to retention and equity outcomes.
Comments
No comments yet.