University On-Screen Marking Implementation Checklist

The pipeline

1. 01MaskFictitious number. Flap torn off.
2. 02ScanOverhead or high-speed. Books may be opened.
3. 03CloudPrivate server. Copy, pages, quality.
4. 04AgencyEvery copy reviewed page by page.
5. 05PacketsApproved books assigned to examiners.
6. 06RevalUnmarked original. Marks stored separately.

Use this as a working document in an examination cell. Answer the questions in writing. If a vendor cannot answer them, that is information. Full conceptual background: Complete Guide to On-Screen Marking & Digital Evaluation. Product description if you are evaluating Avid: Avid OSM.

Planning

• Examination volume. How many answer books, courses and sessions are in the first OSM season? What is the peak day?
• Evaluation model. Centralised university evaluation, college-level marking, mixed? Who is the awarding authority?
• Examiner requirements. How many examiners, paper-wise eligibility, existing empanelment lists, training time before the calendar freeze?
• Pilot vs full session. Which paper is the rehearsal? What does “success” mean (scan reject rate, days to close evaluation, grievance volume)?
• Languages and papers. Which scripts appear in student answers? If AI is contemplated, coverage is confirmed per paper — not assumed.

Scanning

• Image quality. Minimum resolution, colour policy, handling of graphs, maps and supplementary sheets.
• Indexing. How is each copy uniquely identified after fictitious numbering / flap tear-off?
• Script identification. Who confirms page count and sequence before assignment?
• Recapture. What happens to a rejected scan? Who owns the physical book at that moment?
• Capacity. Overhead vs high-speed scanners; opening/unstapling books; agency vs in-house staff.

Security

• Access control. Which roles see unassigned copies, assigned packets, bulk exports?
• Encryption where applicable. In transit and at rest — and who holds keys / hosting?
• Audit trails. Assignment changes, mark edits, downloads, AI runs.
• Data retention. How long unmarked originals and mark layers are kept; who may delete them.
• Anonymity. Is student identity absent from the examiner view as intended?

Narrative notes: digital evaluation security.

Evaluation

• Examiner allocation. Packets, paper-wise limits, what the exam cell sees in realtime.
• Question-wise marking. Compulsory parts, optional questions, best-of — frozen before assignment rush.
• Moderation. Sampling, second marking, or scaling — as your statute requires, not as a vendor template.
• Review. How an examiner raises a defective scan without stalling the packet silently.

Revaluation

• Script retrieval. Unmarked original vs a copy already annotated.
• Reassignment. Different examiner; conflict rules; second revaluation if your ordinance has it.
• Audit trail. Each look recorded as a controlled update to the result, not an informal overwrite.

Results

• Marks integration. File, API, or named ERP/SIS? Who signs off the gazette extract?
• ERP / result workflow. If using campus ERP, which workspace receives OSM output? Avid’s ERP editions are described separately on University ERP and College ERP — only relevant if those products are in the same programme.

AI

• Human oversight. Can AI be disabled for a paper or a season?
• Explainability. Is a reason shown with a suggested mark?
• Examiner override. Can every suggested field be changed before submit?
• Governance. Written policy: AI does not autonomously finalise marks. Who is accountable in a grievance?

Model explanation: AI-assisted evaluation, human-in-the-loop.

What this checklist is not

It is not a claim that Avid has implemented OSM at a named list of universities. It is not a statutory substitute. It is not a request for a backlink. If you need a walkthrough of Avid’s own software, book an OSM demo — that is optional.

FAQ