# Agent context: Students Leaving Before They Finish Their Degree

Source: https://aiplaybooks.app/industries/higher-education/student-success/student-attrition
AI suitability: conditional

## Objective

Raise the share of admitted students who continue and graduate by finding students who are going off track early enough for an adviser or tutor to help, using alerts tied to specific causes first and any risk model only to order outreach, without labelling students by who they are.

## Constraints

- A risk score orders an adviser's outreach and nothing else. Never use it to decide admission, financial aid, academic standing, probation, a change of major, or to encourage a student to leave.
- Never display a risk score or a "high risk" label on the student's record, to faculty, or to the student, and never show predicted risk by major.
- Do not use race, ethnicity, sex, income, or first-generation status as model inputs, a policy choice since the research on race inputs is mixed. Hold them separately and use them only to measure who is flagged, contacted, and retained by group, at the cutoff actually used.
- Removing demographic inputs does not remove bias, because grades and other records carry it, and the same holds for engagement flags. Measure the flagged and missed rates per group rather than assuming fairness.
- Train only on information known at the time of each prediction, and count students who transferred and graduated elsewhere as completers, not dropouts.
- Evaluate any score against a randomised or rotating holdout of outreach in the adviser's usual order. Do not claim effect from comparisons where students who stayed longer simply received more of the intervention.
- In the United States, education records and anything derived from them fall under FERPA, and a vendor may receive them only under the institution's direct control. In the EU, an AI system that steers a student's learning or assigns them to programmes may be high-risk under Annex III of the AI Act, with obligations applying from 2 December 2027.

## Inputs

- Student information system records by term, as known at the time, including registration, major, credits, GPA, and holds.
- Degree maps per programme with required courses, their recommended terms, and courses whose grades predict success in the major.
- Student accounts balances, holds, and the dates a balance would drop a student from classes.
- Learning management system activity per course and week, with course cohort averages and the teaching calendar.
- Advising and tutor contact records with each alert, who received it, whether the student was contacted, and the outcome.
- Continuation and graduation outcomes, including from the National Student Clearinghouse or equivalent, and protected characteristics held apart from model inputs.

## Outputs

- Specific alerts, each naming its trigger, such as a course off the degree map or a balance due to drop the student, routed to the student's own adviser with a response deadline.
- Weekly engagement flags to personal tutors for students far below their course cohort.
- Where justified, an ordering of each adviser's caseload for outreach, with no score visible on the student's record.
- A termly report of alerts raised, contacted, and resolved, and of flagged and missed rates and continuation by group.

## Success criteria

- Continuation and graduation rates rise for the cohorts served, measured against a comparison that accounts for changes in who enrolled.
- Most alerts lead to contact within the agreed time, and credits taken off the degree map and drops over small balances fall.
- Any score-ordered outreach beats the holdout in continuation, or it is retired.
- The gap between groups in the share of at-risk students reached does not widen, and no group is flagged at a rate its outcomes do not justify.

## Known pitfalls

- Buying a predictive platform without the advisers to act on its alerts, so a list of at-risk students grows while caseloads stay too large to reach them.
- Showing a "high risk" label on the student's record, where it shapes every adviser's first impression, as the investigation of EAB Navigate described.
- Using predicted risk by major in a conversation about which major a student should take, which experts warned would push Black students out of science and maths.
- Assuming that leaving race out of a model removes its bias, when research found accuracy almost unchanged without demographic inputs and students of colour still flagged more often, because GPA and other inputs carry the same information.
- Claiming an intervention works because the students who received it more often stayed longer, the flaw critics found in Purdue's Course Signals retention claim.
- Identifying students likely to drop out in order to encourage them to leave before they count in retention figures, which the president of Mount St. Mary's University proposed with a freshman survey, in a controversy that ended with his resignation in 2016.

## When not to use AI here

- The institution cannot staff timely outreach, so a better ranked list changes nothing for the student.
- The losses come from causes a rule can find exactly, such as courses off the degree map, missing registration, or small balances that trigger a drop.
- The score would be used to decide admission, aid, academic standing, or a change of major, or to encourage a student to leave.
- The institution cannot measure who the model flags and misses by race, income, and first-generation status, or will not act when the rates diverge.
- Most students who leave share an environment, such as a campus, programme, or schedule, which research on a school dropout warning system in Wisconsin suggests can be targeted as efficiently as with individual scores.

## Candidate approaches

- Degree map, registration, and balance alerts routed to a named adviser (primary): Fixed alerts that fire when a student registers off their degree map, earns a low grade in a course that predicts success in their major, has not registered for next term, or is about to be dropped over a small balance, each sent to the student's own adviser to act on within a set time.
- Course engagement screening for personal tutors (discovery): A weekly screen of each student's activity in the learning management system and other campus systems against their course cohort, which flags students whose engagement has fallen far below their peers to their personal tutor or a support team, using no demographic inputs.
- Persistence risk score to order advisers' outreach (supporting): A model trained on past students' records that estimates each current student's chance of not returning next term, used only to order the outreach list within an adviser's caseload, built without race, measured by group, and never shown as a label on the student's record.
- Course chatbot outreach for large gateway courses (supporting): A text-message assistant for a large introductory course that sends reminders about assignments, exams, and help sessions, answers routine questions about the course from material the instructors approve, and passes anything personal or difficult to the instructor or an adviser.

## Human oversight

An academic adviser or personal tutor decides whether and how to contact a student and what to recommend, using the alert as a prompt alongside what they know of the student. No change of major, withdrawal, probation, financial decision, or admission decision is ever taken because of a risk score. Faculty decide grades and academic standing. The student success leadership owns any predictive model, reviews who it flags and misses by group at least once a year, and stops it if outreach based on it does not beat outreach without it.

## Sources

- [Academic Early Alert and Intervention: Why Academic Advisors Are Best Suited to Intervene with At-Risk Students](https://nacada.ksu.edu/Resources/Academic-Advising-Today/View-Articles/Academic-Early-Alert-and-Intervention-Why-Academic-Advisors-Are-Best-Suited-to-Intervene-with-At-Risk-Students.aspx), NACADA Academic Advising Today, University of South Carolina advisers, 2020, practitioner
- [Proactive Advising: A Playbook](https://proactiveadvising.theuia.org/assets/uia-proactive-advising-playbook.pdf), University Innovation Alliance, practitioner
- [What ASU learned from Proactive Advising: Lisa McIntyre](https://www.youtube.com/watch?v=dhKJxmlMtK8), University Innovation Alliance, YouTube, 2019, practitioner
- [Why the Course Signals Math Does Not Add Up](https://hapgood.us/2013/09/26/why-the-course-signals-math-does-not-add-up/), Mike Caulfield, Hapgood, 2013, practitioner
- [MAAPS Advising Experiment: Evaluation Findings after Six Years](https://sr.ithaka.org/wp-content/uploads/2023/05/SR_Report_MAAPS_EvaluationFindings_6Years_05162023.pdf), Ithaka S+R, 2023, research
- [Big Data on Campus](https://www.educationnext.org/big-data-on-campus-putting-predictive-analytics-to-the-test/), Education Next, Bird, Castleman, Mabel, and Song, 2021, research
- [Are algorithms biased in education? Exploring racial bias in predicting community college student success](https://doi.org/10.1002/pam.22569), Journal of Policy Analysis and Management, Bird, Castleman, and Song, 2025, research
- [Difficult Lessons on Social Prediction from Wisconsin Public Schools](https://arxiv.org/abs/2304.06205), Perdomo, Britton, Hardt, and Abebe, ACM Conference on Fairness, Accountability, and Transparency, 2025, research
- [Discretion in the Loop: Human Expertise in Algorithm-Assisted College Advising](https://arxiv.org/abs/2505.13325), Schechtman, Brandon, Stafford, Li, and Liu, arXiv, 2025, research
- [Let's Chat: Leveraging Chatbot Outreach for Improved Course Performance](https://edworkingpapers.com/sites/default/files/ai24-564_v2.pdf), Meyer, Page, and others, Annenberg Institute EdWorkingPaper, 2024, research
- [Building a Pathway to Student Success at Georgia State University](https://sr.ithaka.org/publications/building-a-pathway-to-student-success-at-georgia-state-university/), Ithaka S+R, 2015, case-study
- [Creating Positive Impacts On Student Engagement: Nottingham Trent University](https://kortext.com/stream/wp-content/uploads/2024/08/Creating-positive-impacts-Case-Study-NTU.pdf), Kortext, the vendor of the dashboard, 2024, case-study
- [Major Universities Are Using Race as a "High Impact Predictor" of Student Success](https://themarkup.org/machine-learning/2021/03/02/major-universities-are-using-race-as-a-high-impact-predictor-of-student-success), The Markup, 2021, case-study
- [Texas A&M Drops "Race" from Student Risk Algorithm Following Markup Investigation](https://themarkup.org/machine-learning/2021/03/30/texas-am-drops-race-from-student-risk-algorithm-following-markup-investigation), The Markup, 2021, case-study
- [False Alarm: How Wisconsin Uses Race and Income to Label Students "High Risk"](https://themarkup.org/machine-learning/2023/04/27/false-alarm-how-wisconsin-uses-race-and-income-to-label-students-high-risk), The Markup, 2023, case-study
- [President quits at Mount St. Mary's](https://www.insidehighered.com/news/2016/03/01/president-quits-mount-st-marys), Inside Higher Ed, 2016, case-study
- [Numerical thresholds for condition B3](https://www.officeforstudents.org.uk/for-providers/quality-and-standards/how-we-regulate-student-outcomes/numerical-thresholds-for-condition-b3/), Office for Students, 2023, regulation
- [34 CFR Part 99, Family Educational Rights and Privacy](https://www.ecfr.gov/current/title-34/subtitle-A/part-99), U.S. Department of Education, Electronic Code of Federal Regulations, regulation
- [Regulation (EU) 2024/1689, the Artificial Intelligence Act](https://eur-lex.europa.eu/eli/reg/2024/1689/oj/eng), Official Journal of the European Union, regulation
- [Regulation (EU) 2026/1744, the Digital Omnibus on AI](https://eur-lex.europa.eu/eli/reg/2026/1744/oj/eng), Official Journal of the European Union, 2026, regulation