An online proctoring report helps to verify the candidate's identity, timestamps, alerts, technical issues, and flags, ensuring smooth online exam monitoring to maintain exam integrity. The exam team can analyze the report and understand what happened during the session to take necessary steps, maintaining exam integrity.
An online proctoring report helps exam teams to supervise remote sessions and understand what happened during the activity, empowering them to analyze the challenges and take required steps. Depending on the system, it may include candidate identity checks, activity timestamps, AI-generated alerts, video and audio evidence. Also, faculty and invigilators can review technical interruptions during or after the assessment.
The proctoring report may include alerts or proctoring flags that were generated during the session when a potential violation was detected by the online exam monitoring software. The exam team needs to verify underlying evidence, check for technical challenges, and document how each concern was resolved.
How should the Exam Team Handle a Flag Event?
During an online exam, an AI proctoring system may generate false alerts/flags. Thus, a human review in online exam helps to understand the event and take necessary actions.
- Step 1: Identify the event: Read the alert description, timestamps, and relevant exam rule.
- Step 2: Verify the evidence – Examine the available recording and audio, and the event logs.
- Step 3: Review - Apply consistent criteria to determine whether the event is explained, inconclusive, or warrants further investigation.
- Step 4: Record the Outcome - Document the reasoning, any action taken, and the appropriate follow-up or appeal route.
Conclusion
Online proctoring reports can help exam teams investigate irregularities, review session events, and document assessment decisions. Their value, however, depends on the quality of the underlying data and the process used to interpret it.
Before acting on a report, verify identity checks, timestamps, supporting evidence, technical incidents, reviewer decisions, and privacy controls. Most importantly, treat automated flags as prompts for review—not as automatic proof of misconduct.
