{"resourceId":"coventry-early-student-support-ai-attendance-aula-20261007","versions":[{"version":"external-9dfeaf4ea0d41638eb29e9244b87ad348df5a6c9cdd8ac9cde8311bd59b1c0ac","resource":{"id":"coventry-early-student-support-ai-attendance-aula-20261007","title":"Coventry Early Student Support AI flags attendance and Aula activity for staff outreach after evaluation on more than 9,000 student records","organization":"Coventry University (Professor James Brusey, Research Centre for Computational Science and Mathematical Modelling; Andrew Turner, Vice-Provost and Deputy Vice-Chancellor Education)","sector":"Higher education student success / early alert, retention and advising support","geography":"United Kingdom (Coventry University)","publishedAt":"Coventry University news dated Wednesday 7 October 2026","publicationDate":"2026-10-07","eventDate":null,"sourceName":"Coventry University News","sourceLabel":"Institutional news interview with project lead and vice-provost describing an early-stage AI early-support project and evaluation scale","sourceUrl":"https://www.coventry.ac.uk/news/2026/coventry-university-ai-tool-students-support/","evidenceClass":"independent-reporting","outcomeClass":"emerging","topics":["operating-model","data-security","accessibility-workforce"],"finding":"Coventry University's Early Student Support project uses AI to analyse attendance records and activity on Aula, the university's learning platform, so staff can spot potential issues sooner. Led by Professor James Brusey and delivered through the AI Adoption Lab, the work has been evaluated using more than 9,000 student records across the first two teaching blocks of 2025-26, with additional historical datasets under review. Students are not contacted by AI; when the system identifies someone who may need support, a staff member reaches out to discuss circumstances and help. Vice-Provost Andrew Turner said the technology is not there to replace people or make decisions about students, and that staff remain responsible for understanding circumstances and determining support. Brusey described the project as early-stage with encouraging initial results but did not publish precision, recall, fairness or retention-effect figures in the article.","sledRelevance":"Dated 7 October 2026 and not in the archive at check_sources, it is a concrete HE retention architecture: LMS and attendance signals, human-only outreach, and an explicit non-decision role for AI. Useful contrast to tutoring tools that act directly on students.","evidence":"University news with named leaders; stated evaluation on more than 9,000 student records in teaching blocks 1–2 of 2025-26; process description of staff-only contact. No published performance metrics or outcome deltas in the article.","architectureImplications":"AI analyses attendance plus Aula engagement features to surface candidates for human follow-up; no automated student messaging or decisioning described.","governanceImplications":"Interpretation: keep early-alert AI in an advisory role with named staff ownership of outreach and support decisions; publish how students are selected and how to contest flags.","securityPrivacyImplications":"Combining attendance and LMS activity is sensitive; the article does not detail retention, access controls or student notice, which campuses must specify before scale.","caveats":"Institutional news, not an independent evaluation; 'encouraging' results without effect sizes; early stage; UK setting may limit transfer of data-protection and advising models.","streamIds":["student-success"],"roles":{"sales":"The framing point is AI that advises staff, not students. Coventry's Early Student Support project scores attendance and Aula activity so humans can intervene earlier, after work on more than 9,000 student records, and AI never contacts students. Use it with retention and advising leaders designing early-alert workflows. Bound it: university news, early stage, and no published precision or retention effects.","engineering":"For engineering, the framing point is advisory analytics on attendance plus LMS activity with no automated outreach. Build flag queues for named staff, log who acted and when, and hard-block any student-facing message path from the risk model. Confirm FERPA or GDPR notice, retention limits and least-privilege access before connecting SIS attendance feeds and Aula-like LMS features to the scorer in production.","delivery":"Delivery's framing point is staffing the loop. Assign owners for each flag, set response-time targets, train advisers on how the score is produced and its error modes, and measure outreach completion plus later progression—not model accuracy alone. Review fairness of who gets flagged before expanding colleges. Do not declare success until human follow-through and student outcomes are tracked across a full teaching block."},"roleDetails":{"sales":{"customerProblem":"Advising teams cannot scan attendance and LMS activity at scale, so at-risk students are contacted late or unevenly.","stakeholders":["Vice-provost or associate provost for student success","Retention / advising leadership","Success coaches","Institutional research","Data protection officer"],"discoveryQuestions":["Who is allowed to contact a flagged student, and how soon?","What data sources feed the model, and have students been notified?","Which outcomes will count as success besides model accuracy?"],"valueHypothesis":"Advisory early-alert scoring may focus limited adviser time earlier; Coventry has not yet published outcome effects.","potentialEngagement":"An early-alert design review: data inventory, human-in-the-loop workflow and evaluation plan on a pilot college.","unsupportedClaims":["That the Coventry tool improves retention","That AI contact is unnecessary in all contexts","That 9,000 records prove accuracy"]},"engineering":{"fit":"Universities with attendance and LMS activity data and staff ready to own outreach.","architectureIntegration":"Batch or near-real-time scoring of attendance and LMS features into a staff-only work queue; no student-facing chatbot path from the risk model.","prerequisites":["Attendance and LMS data agreements","Staff identity and case-management hooks","Audit log of flags and actions"],"constraints":["Early-stage evaluation","UK legal context differs from U.S. FERPA practice","Risk of biased flags without fairness checks"],"security":"Restrict flag visibility to assigned staff, minimize retained features, and document lawful basis / FERPA school-official rationale.","proofOfValue":["Outreach completed within the target window","False-positive burden acceptable to advisers","Fairness audit across student groups before scale"]},"delivery":{"work":"Stand up a staff-only early-alert queue, train advisers, and run a time-boxed pilot with progression measures.","dependencies":["Adviser capacity","IR evaluation design","Data protection approval"],"ownership":"Student success / advising owns outreach decisions; data science owns the model; DPO owns privacy controls.","skillsAdoption":"Adviser training on interpreting flags without treating them as ground truth.","governanceCheckpoints":["Privacy review","Fairness check on pilot flags","Go/no-go after teaching block"],"acceptanceCriteria":["Zero automated student contacts from the model","Documented owner and response SLA for each flag","Pilot report includes outreach completion and next-term progression"],"risks":["Alert fatigue","Stigmatizing contact","Scaling before fairness and outcome evidence"]}},"retrievedAt":"2026-10-09T03:30:01Z","enrichedAt":"2026-10-09T03:33:56Z","enrichmentBasis":"retrieved source","accessibilityWorkforceImplications":"Staff capacity to hold supportive conversations is the scarce resource the tool is meant to focus, not replace.","procurementImplications":"Buy or build early-alert features only with contractual human-in-the-loop requirements and auditability of flags.","operatingModelImplications":"Define who receives flags, expected outreach time, and how success coaches or advisers close the loop.","sourceVerification":{"openedUrl":"https://www.coventry.ac.uk/news/2026/coventry-university-ai-tool-students-support/","referenceExcerpt":"Students are not contacted by AI however - if the system identifies someone who may need support, a member of staff reaches out","promptVersion":"sled-research-v4.0","model":null,"basis":"agent-reported inspection"}}}]}