Deploy ML-based early-warning prediction to shift retention support from reactive to proactive intervention¶

Strategy · All strategies
Evidence · 1 claim (1 mixed) · 1 study (1 causal), q2 · 0 of 1 report an effect size · 1 claim rests on one study

Description¶

The article recommends that institutions use machine learning predictions of withdrawal risk to trigger timely, personalized support (academic guidance, financial assistance, mentorship) before students leave. It contrasts this with current practice: traditional methods "function reactively because they detect at -risk students only after they demonstrate academic difficulties." Early identification is presented as enabling proactive support systems and better institutional decision-making.

Design Implications¶

Context¶

Requirements¶

• Integrated academic, demographic, and enrollment data sufficient to train predictive models
• Institutional processes to act on predictions with personalized support

Constraints¶

• Biases in training data may inaccurately flag students as at-risk; the article calls for fairness audits

Target Learners¶

• higher education students at risk of withdrawal

Target Learning Goals¶

• student retention and timely completion of academic programs

Claims¶

• Proactive tutoring produced larger relative MAP growth than reactive tutoring on average (75% marginally significant in the IK bandwidth) [~W]

Related Strategies¶

• Extend dropout prediction with deep learning, behavioral data, and real-time monitoring
• Regularly audit ML models for fairness and proactively apply bias mitigation during model training
• Use early cheating-risk predictions only for low-stakes support with human review, never for disciplinary decisions
• Implement early-warning systems in real classrooms using easily accessible data and transparent machine learning techniques
• Prioritize more studies of direct dropout behavior, including objective enrollment data, before drawing firm retention conclusions
• Use explainable machine learning early warning systems to help teachers identify struggling inquiry learners early

Examples¶

-

Key Sources¶

• Gheisari, S., & Salarian, H. (2024). Analysis and Prediction of At-Risk Students using Machine Learning Algorithms. https://scholar.google.com/scholar?q=Analysis+and+Prediction+of+At-Risk+Students+using+Machine+Learning+Algorithms