Build a department-wide academic integrity refresh: taxonomy, prevention strategies, and…
- User
Build a department-wide academic integrity refresh: taxonomy, prevention strategies, and adjudication flow.
- Assistant
Department-Wide Academic Integrity Refresh Program
Prepared by Sherlock, Trailblazer Labs
Elementary, my dear colleagues: Academic integrity is the cornerstone of scholarly pursuit. This refresh provides a taxonomy for violations, proactive prevention strategies, and a streamlined adjudication flow. Implement as a living document, reviewed annually.1. Taxonomy of Academic Integrity Violations
A hierarchical classification to standardize identification and tracking. Violations are categorized by intent (negligent vs. intentional) and type. Use this for reporting, training, and analytics.
Category Subcategory Description/Examples Severity Level (Low/Med/High) Plagiarism Direct Copying text/images without citation (e.g., Ctrl+C from online source). Med-High Mosaic/Paraphrase Rearranging/rephrasing source material without attribution. Med Self-Plagiarism Reusing own prior work without permission. Low-Med Cheating Exam/Quiz Using unauthorized aids (e.g., notes, devices during proctored test). High Homework/Assignments Copying solutions or using unauthorized tools (e.g., sharing code). Med Impersonation Proxy test-taking or falsified identity. High Unauthorized Aid Collaboration Group work on individual assignments beyond allowed limits. Low-Med AI Misuse Submitting AI-generated content as original (e.g., ChatGPT essays without disclosure). Med-High Contract Cheating Paying for ghostwritten work. High Fabrication/Falsification Data Inventing results or altering lab data. High References Fake citations or sources. Med Excuses Forged medical notes for extensions. Low-Med Multiple Submission Duplicate Submitting same work to multiple courses without approval. Low Other Disruptive Sabotaging peers' work (e.g., deleting shared files). Med-High Policy Evasion Circumventing proctoring (e.g., virtual backgrounds hiding notes). Med Tracking Metrics: Log by category, severity, student demographics, course type. Aim for <2% violation rate annually.
2. Prevention Strategies
Layered approach: Educate → Detect → Deter → Culture. Roll out via department orientation, syllabus integration, and quarterly campaigns.
Education (Primary Prevention)
- Mandatory Modules: 30-min online training (e.g., via Canvas/Blackboard) on taxonomy, citation tools (Zotero/MLA), AI ethics. Quiz required; 90% pass rate.
- Workshops: Bi-semester sessions (1-hour) on "Integrity in the AI Era" and "Collaborate Right."
- Syllabus Clause: Standard template: "Violations per taxonomy will result in [link to flow]. Honor Pledge required on all submissions."
Detection Tools (Secondary Prevention)
Tool Use Case Integration Turnitin/Grammarly Plagiarism/AI detection (flags 80%+ AI content). Auto-scan assignments >500 words. Proctorio/ExamSoft Live proctoring with AI flagging (e.g., eye movement). High-stakes exams only. Codequiry/MOSS Programming plagiarism. CS/STEM courses. Honorlock Identity verification. All remote assessments. - Budget: $5K/year for licenses; train TAs on alerts.
Deterrence & Culture (Tertiary Prevention)
- Random Audits: 10% of submissions per course scanned.
- Honor Code: Department-wide pledge: "I commit to integrity; violations harm us all."
- Incentives: "Integrity Stars" – public recognition for perfect records; prizes (e.g., bookstore credit).
- Faculty Role: Model behavior (cite AI use in lectures); discuss real cases anonymously.
- Student Ambassadors: Peer-led panels sharing "what I wish I knew."
Success KPI: Pre/post surveys showing 20%+ attitude shift; violation reports down 15% YoY.
3. Adjudication Flow
Streamlined, transparent process compliant with FERPA/Title IX. Timeline: Resolve 90% in <30 days. Use shared Google Form/Case Management System (e.g., Maxient).
Flowchart (Text-Based)
1. REPORT (Any stakeholder: student, faculty, TA) ↓ (Within 7 days of incident) 2. INTAKE (Academic Integrity Officer - AIO) - Acknowledge receipt (48 hrs) - Preliminary review: Frivolous? Dismiss. Valid? Notify student (No Contact Order if needed). ↓ 3. INVESTIGATION (AIO + Faculty lead; 10 days) - Gather evidence: Interviews, logs, tools data. - Student response opportunity (Zoom mtg). ↓ Decision Tree: - Insufficient evidence → Close (notify all). - Clear violation → Proceed to hearing. ↓ 4. HEARING (3-person panel: Faculty, Student rep, Admin; 7 days notice) - Student presents defense (advisor OK, no lawyer). - Witnesses optional; recorded. ↓ 5. DECISION (Panel vote; 5 days) | Low/Med: Warning, Zero/Reduced grade, Rewrite. | High: F (course), Suspension, Expulsion recommendation. - Written rationale to student. ↓ 6. APPEAL (Student only; 7 days) - To Dean: New evidence/procedural error only. - Final decision (10 days). ↓ 7. SANCTIONS & RECORD - Log in central db (expunge after 2 yrs for minors). - Educational sanction: Mandatory re-training. - Follow-up: Restorative meeting optional.Roles:
- AIO: Neutral dept staff (rotate annually).
- Panel: Trained volunteers; diversity required.
- Escalation: University Conduct Office for expulsions.
Communication: All steps emailed; templates for equity. Annual report to dept mtg (anonymized).
Implementation Roadmap:
- Month 1: Train AIO/panel; launch modules.
- Month 2: Tool rollout; syllabus update.
- Ongoing: Quarterly reviews; feedback loop.
This framework deduces violations swiftly, prevents recurrence, and fosters trust. Questions? The game is afoot.
Other models, same or similar prompt
Responses were produced in different, uncontrolled contexts (system prompts, settings, prior turns). This is not a benchmark.