Remote assessment integrity guide

Complete Guide to Online Proctoring

Understand how online proctoring supports remote assessment integrity through identity checks, secure test delivery, live or recorded monitoring, automated signals, evidence review, candidate communication, accessibility, privacy controls, incident handling, and accountable human decisions.

Discuss Online Proctoring Explore the Guide
Identity and session controls
Evidence-based review
Candidate safeguards
Candidate completing a secure online assessment on a laptop
Secure remote assessment requires more than webcam access A complete online proctoring process includes clear candidate communication, technology checks, proportionate monitoring, evidence review, privacy controls, accessibility support, and documented decision responsibility.

Online proctoring fundamentals

What online proctoring does and what it does not do

Online proctoring supports remote assessment supervision and evidence collection. It should not replace assessment design, candidate support, investigation, accountable judgement, or clear policies.

01
What online proctoring can support

Controls and records across a remote assessment session

Depending on the selected model and configuration, online proctoring may assist with identity confirmation, secure test delivery, webcam or screen monitoring, event recording, live supervision, and post-session review.

Candidate identity workflow

Collect and review approved identity evidence before the test.

Assessment environment checks

Confirm permitted devices, browser readiness, and session rules.

Session event records

Preserve relevant timestamps, interruptions, permission events, and monitoring evidence.

Human review workflow

Support investigation and documented decisions after the exam.

02
What online proctoring cannot establish alone

Monitoring information still requires context and interpretation

A face temporarily leaving the frame, an unexpected sound, a browser interruption, or a second device signal may have several explanations. Technology should not convert uncertain events into automatic misconduct conclusions.

Candidate intent

A monitoring event does not independently establish why it occurred.

Assessment validity

Proctoring does not make weak questions or scoring rules valid.

Final disciplinary decision

Significant outcomes require documented policy and accountable review.

Fair candidate access

Accessibility, technology, privacy, and support require separate planning.

Online proctoring models

Choose a proctoring model that matches the assessment context

Live, recorded, automated, and hybrid proctoring distribute monitoring and review work differently. Selection should consider risk, scale, candidate needs, cost, privacy, review capacity, and the consequences of the assessment.

LP
Live proctoring

A proctor supervises the candidate during the assessment

The proctor may verify identity, communicate with the candidate, observe the session, respond to incidents, and document relevant events in real time.

Useful consideration Confirm proctor training, candidate-to-proctor ratio, escalation, language coverage, and incident documentation.
RR
Record and review

The assessment is recorded and reviewed after completion

Webcam, microphone, screen, browser, or session information may be captured for later review by an authorised person.

Useful consideration Define evidence retention, reviewer workload, review criteria, response time, and candidate communication.
AP
Automated proctoring

Technology identifies events or patterns for review

Automated features may generate timestamps, alerts, classifications, confidence indicators, or prioritised review queues based on configured signals.

Useful consideration Validate signal quality, false alerts, language and device performance, limitations, and human-review requirements.
HP
Hybrid proctoring

Automated support and human supervision operate together

Technology may organise evidence or identify events while trained people supervise, investigate, review, communicate, and make decisions.

Useful consideration Document the boundary between automated signals, proctor actions, reviewer findings, and final decisions.

End-to-end proctoring workflow

Build a complete session dossier from invitation to final review

The proctoring process begins before the candidate opens the test and continues after submission. Each stage should have documented rules, evidence, ownership, exceptions, and candidate communication.

Preparation stage

Configure the assessment and communicate requirements

Define the proctoring model, allowed resources, identity method, technology requirements, environment rules, support routes, accommodations, and incident process.

Publish clear candidate instructions.
Test devices, browser, camera, microphone, and network.
Confirm support and accommodation contacts.
01
Evidence to retain

Configuration and communication record

Preserve the active assessment version, rules, candidate notice, technical requirements, permissions, and approved exceptions.

Identity stage

Confirm the approved candidate identity process

The identity workflow may involve login credentials, approved identification evidence, a live proctor, a photograph, or another authorised verification method.

Explain what identity information is required.
Restrict access to identity records.
Provide a process for unclear or failed verification.
02
Evidence to retain

Verification result and exception record

Record the outcome without exposing more personal information than necessary to assessment reviewers.

Environment stage

Confirm the permitted assessment setting

Depending on policy, candidates may be asked to confirm their surroundings, workspace, allowed materials, browser state, or connected devices.

Use proportionate and clearly explained checks.
Avoid collecting unrelated private information.
Provide alternatives where approved.
03
Evidence to retain

Environment readiness and exception notes

Store only information relevant to the approved proctoring purpose and investigation process.

Assessment stage

Monitor the session and protect assessment continuity

Record approved session events, support interactions, interruptions, connectivity changes, permission failures, proctor instructions, and candidate responses.

Distinguish technical events from candidate behaviour.
Provide a visible way to request help.
Preserve timestamps for significant events.
04
Evidence to retain

Session event and support record

Maintain relevant logs, recordings, proctor notes, technical events, candidate messages, and submission status.

Review stage

Examine relevant evidence before reaching a conclusion

Reviewers should consider the complete session, candidate explanation, technical information, assessment rules, event context, proctor notes, and limitations of automated signals.

Use documented review criteria.
Escalate uncertain or consequential cases.
Record the evidence supporting the decision.
05
Evidence to retain

Reviewer findings and decision record

Document who reviewed the session, what evidence was considered, any limitations, the conclusion, and the next action.

Signals and conclusions

Separate monitoring events from final integrity decisions

Online proctoring may produce useful evidence, but individual signals can be incomplete, ambiguous, or affected by technology. Review the event in context before assigning meaning.

Possible session signals

Events that may require review

Signals can help organise review work, identify interruptions, or direct attention to specific timestamps.

Visual event Candidate or face temporarily leaves the camera frame.
Audio event Unexpected speech, background noise, or microphone interruption.
Browser event Window change, permission change, or browser focus event.
Device event Camera, microphone, screen capture, or network status changes.
Responsible review questions

Context required before reaching a conclusion

Reviewers should determine what occurred, whether the evidence is reliable, and whether an alternative explanation is supported.

Technical context Was there a documented connectivity, browser, device, or permission issue?
Candidate context Did the candidate request help or provide a relevant explanation?
Assessment context Which rule applies and was it communicated clearly?
Evidence context Does the complete session support the same interpretation?

Candidate experience

Design online proctoring around clear participation

Candidates should understand what the assessment requires, why proctoring is used, what information is collected, which technology is needed, where to obtain support, and how incidents or accommodations are handled.

Before the exam

Explain the process in direct language

Include equipment, permissions, room requirements, identity steps, allowed materials, expected duration, and support.

Accessibility

Provide a confidential accommodation route

Test alternative arrangements and avoid treating accommodation information as assessment evidence.

During the exam

Make support visible and responsive

Candidates should know how to report permission, network, device, recording, login, or submission problems.

After the exam

Confirm submission and explain next steps

Provide appropriate confirmation, support escalation, incident information, and process communication.

Students preparing for secure online examinations in a learning environment
Candidate-centred proctoring Security controls should remain proportionate, understandable, accessible, supportable, and connected to the assessment purpose.

Technology readiness

Test the complete proctored assessment environment

Technology testing should cover candidate preparation, session permissions, monitoring, interruptions, recovery, support, submission, evidence access, and reviewer workflows.

DV
Device readiness

Confirm supported equipment and permissions

Define the supported desktop or mobile environment and test the required camera, microphone, screen, and browser permissions.

Supported operating systems and browsers
Camera, microphone, and speaker readiness
Screen dimensions and device limitations
NW
Network readiness

Prepare for unstable or interrupted connectivity

Define what happens when a recording pauses, a session disconnects, an upload fails, or the candidate needs to resume.

Pre-exam connectivity test
Reconnection and resume behavior
Upload and submission recovery
BR
Browser controls

Configure restrictions without creating avoidable failure

Test secure browser requirements, focus monitoring, tab behavior, pop-up restrictions, clipboard settings, and permitted resources.

Browser compatibility and installation
Allowed links, tools, and resources
Permission and extension conflicts
SP
Support readiness

Connect technical events with human support

Provide visible help routes, documented escalation, event records, candidate communication, and clear retry or rescheduling rules.

Candidate help contact
Incident ownership and response time
Retry, pause, and reschedule procedures

Illustrative review interface

Organise session evidence without replacing reviewer judgement

A review interface can help authorised users locate events, technical records, proctor notes, candidate messages, recordings, and decisions. The example values below are illustrative.

Online Proctoring Session Review Illustrative view
Session review overview

Evidence, technical events, and reviewer actions

Current assessment window
Sessions completed 248 Illustrative count
Review queue 19 Illustrative count
Technical cases 11 Illustrative count
Reviews closed 92% Illustrative rate
Illustrative session events

Evidence requiring contextual review

09:14:22 Camera permission temporarily changed Technical
09:32:18 Candidate requested support through session chat Support
09:33:41 Network interruption recorded for 18 seconds Network
10:08:09 Assessment submitted successfully Complete
Illustrative review notes

Context considered before closure

Candidate explanation Candidate reported a temporary browser permission issue.
Technical evidence Event log confirms a short network interruption.
Session continuity Assessment resumed and submitted without content loss.
Illustrative outcome No integrity escalation required after contextual review.

Illustrative values and interface elements demonstrate a review structure. Actual alerts, rules, evidence, access permissions, escalation thresholds, retention periods, and decisions should reflect the assessment context and approved governance framework.

Implementation framework

Introduce online proctoring through five controlled stages

Begin with the assessment purpose and candidate journey before configuring monitoring features. Pilot the process, review evidence, correct weaknesses, and expand only after accountable approval.

01
Define

Document assessment risk and proctoring purpose

Identify the assessment type, candidate population, consequences, allowed resources, monitoring needs, and prohibited uses.

Output: approved proctoring specification
02
Design

Build the candidate and reviewer workflows

Prepare instructions, identity steps, technology checks, accessibility, support, evidence definitions, and review rules.

Output: documented end-to-end process
03
Test

Validate normal and exceptional journeys

Test devices, networks, permissions, interruptions, accommodations, live support, session recovery, evidence access, and deletion.

Output: technical and operational test record
04
Pilot

Run a controlled assessment group

Review completion, candidate feedback, technical incidents, monitoring signals, reviewer workload, false alerts, and support.

Output: pilot findings and corrective actions
05
Govern

Approve, monitor, and reassess the process

Track quality, candidate impact, accessibility, privacy, review consistency, incident outcomes, retention, and assessment changes.

Output: recurring governance and review cycle

Frequently asked questions

Online Proctoring FAQs

Review common questions about remote proctoring models, identity checks, monitoring, automated signals, candidate experience, accessibility, privacy, evidence review, and implementation.

What is online proctoring?
Online proctoring is a remote assessment supervision process that may include identity verification, environment checks, webcam or screen monitoring, live supervision, session recording, automated signals, and post-assessment evidence review.
What are the main types of online proctoring?
Common models include live proctoring, record-and-review proctoring, automated proctoring, and hybrid proctoring that combines technology with live or post-session human review.
Does an automated proctoring alert prove misconduct?
No. An alert identifies an event or pattern that may require review. Reviewers should consider the complete session, technical information, candidate explanation, assessment rules, and limitations before reaching a conclusion.
What should candidates be told before a proctored exam?
Explain the assessment purpose, proctoring model, identity steps, equipment, permissions, environment requirements, allowed materials, recording, data use, expected duration, support, accommodations, and incident process.
How should accessibility be handled in online proctoring?
Review camera and audio requirements, time limits, identity checks, browser controls, instructions, assistive technology, support, and alternative arrangements. Provide a confidential accommodation route and test approved alternatives.
Which online proctoring data may require protection?
Protected information may include identity records, photographs, webcam recordings, audio, screen recordings, browser events, session logs, candidate messages, proctor notes, automated alerts, reviewer findings, and decision records.
How should proctoring evidence be reviewed?
Use documented criteria, inspect the relevant event and surrounding session, consider technical records and candidate explanations, distinguish confirmed facts from assumptions, escalate uncertainty, and record the evidence supporting the outcome.
What should be tested before launching online proctoring?
Test invitations, identity steps, devices, browsers, permissions, cameras, microphones, networks, screen capture, interruptions, retries, support, accommodations, submission, recordings, reviewer access, retention, deletion, and incident procedures.
Which metrics should be reviewed after implementation?
Review invitation delivery, starts, completion, abandonment, technical incidents, support requests, accommodation outcomes, alert rates, review workload, reviewer agreement, candidate feedback, appeals, resolution time, and candidate-group outcomes.
How can CloudTest support online proctoring?
CloudTest can support configurable online assessments, candidate attempt management, secure test delivery, proctoring workflows, monitoring records, result reporting, and review processes. Available capabilities may vary by plan and implementation.
Protect assessment integrity without losing candidate trust

Build online proctoring around evidence, support, and accountability

Select the right proctoring model, explain the process, test the technology, support accessibility, protect candidate data, separate signals from conclusions, require human review, and monitor the complete assessment journey.

Book a Demo Explore CloudTest
Connect with CloudTest

Discover smarter hiring with CloudTest

Explore coding tests, AI interviews, and assessment workflows designed to help modern teams hire faster and with more confidence.

  • AI-powered assessments
  • Faster hiring workflows
  • Trusted by growing teams
Book a Demo