PU
Purpose
Why is proctoring needed?
Connect monitoring controls to a documented assessment risk and decision.
What problem would remain if proctoring were not used?
Online proctoring inquiry guide
Questions to Ask About Online Proctoring
Use these questions to evaluate why online proctoring is needed, which model is appropriate, how candidates will access the process, what information will be monitored, how evidence will be reviewed, which privacy safeguards apply, and where accountable human judgement is required.
Purpose and proportionality
Candidate safeguards
Evidence and accountability
Question routes
Explore online proctoring through six connected conversations
Each conversation answers a different implementation question. Together they help teams evaluate suitability, candidate access, technology, evidence quality, privacy, and accountability.
MO
Model
Which proctoring model fits?
Compare live, recorded, automated, and hybrid approaches.
Which activities require real-time human involvement?
CA
Candidate
Can candidates participate fairly?
Review communication, devices, support, privacy, and accessibility.
What happens when a candidate cannot use the standard journey?
TE
Technology
Will the process remain stable?
Test browsers, permissions, networks, recording, recovery, and submission.
How is an interrupted session recovered and documented?
EV
Evidence
What does each signal mean?
Separate observed events, automated alerts, assumptions, and conclusions.
Which evidence must a reviewer inspect before escalation?
GO
Governance
Who owns each decision?
Assign review, escalation, correction, retention, and monitoring responsibility.
Who can challenge or reverse a proctoring conclusion?
Complete question library
Twenty questions to ask before selecting or expanding online proctoring
Use the answer, supporting evidence, unresolved gaps, responsible owner, and required follow-up to build a documented proctoring decision.
01
Purpose
Why is online proctoring required for this assessment?
Clarify the assessment risk, consequence, decision, and integrity concern that the proctoring process is expected to address.
Listen for
A specific risk linked to the assessment rather than a general
preference for greater surveillance.
Follow-up
Could a less intrusive assessment design or control address the
same risk?
02
Assessment context
How consequential is the assessment outcome?
Consider whether the assessment affects recruitment, academic progression, certification, promotion, licensing, or another significant decision.
Listen for
Clear consequences, review rights, evidence standards, and
decision responsibility.
Follow-up
Does the review process become stronger as the consequence
increases?
03
Proctoring model
Which model is appropriate: live, recorded, automated, or hybrid?
Compare candidate support, monitoring, review workload, scale, privacy, cost, response time, and assessment risk.
Listen for
A model selected through documented requirements rather than a
default platform setting.
Follow-up
Which actions are completed by technology and which require a
person?
04
Identity
What level of candidate identity verification is actually required?
Define the approved identity method, information collected, verification evidence, failure process, access restrictions, and alternative arrangements.
Listen for
Proportionate identity checks that collect only information
necessary for the assessment purpose.
Follow-up
How are unclear, expired, mismatched, or unavailable identity
records handled?
05
Candidate communication
What will candidates be told before entering the assessment?
Candidates should understand the proctoring model, equipment, permissions, recording, identity steps, environment rules, support, accommodations, and data handling.
Listen for
Direct instructions provided early enough for candidates to
prepare or request help.
Follow-up
Has the communication been tested with people unfamiliar with the
platform?
06
Technical readiness
Which devices, browsers, networks, and permissions are supported?
Confirm operating systems, browsers, camera, microphone, screen capture, secure browser requirements, network expectations, and installation steps.
Listen for
Published requirements, pre-assessment testing, and browser-
specific recovery guidance.
Follow-up
What happens when a candidate does not own a supported device?
07
Accessibility
How will candidates request and receive accommodations?
Review camera and audio requirements, identity processes, time limits, assistive technology, environment checks, breaks, support, and alternative arrangements.
Listen for
A confidential, tested process that does not treat accommodation
information as assessment evidence.
Follow-up
Has the complete accommodated journey been tested before launch?
08
Environment checks
Which room, workspace, or resource checks are proportionate?
Define what candidates may be asked to show, why it is necessary, what should not be collected, and which alternatives are available.
Listen for
Checks limited to information relevant to the assessment rules
and approved risk.
Follow-up
How are candidates supported when they cannot access a private
room?
09
Monitoring scope
Which information will be monitored or recorded?
Clarify webcam, microphone, screen, browser, keyboard, device, network, room, identity, interaction, and session-event collection.
Listen for
A defined purpose for each information type and a clear boundary
against unrelated collection.
Follow-up
Could any monitoring feature be disabled without weakening the
documented control?
10
Automated signals
What events can generate automated alerts or classifications?
Identify the monitored event, data source, threshold, confidence, limitations, known errors, supported populations, and required human review.
Listen for
Clear distinction between an alert, an observed event, and a
confirmed policy concern.
Follow-up
How often are alerts dismissed after contextual review?
11
Session interruption
What happens when the network, camera, browser, or recording fails?
Define pause, recovery, reconnect, resume, reschedule, retry, evidence preservation, support, candidate communication, and final submission rules.
Listen for
Technical events treated separately from candidate conduct or
assessment-integrity concerns.
Follow-up
Can reviewers see the technical context when examining an alert?
12
Candidate support
How can candidates obtain help before and during the assessment?
Review support channels, operating hours, language coverage, response time, escalation, event recording, and retry authority.
Listen for
Support that remains visible during identity, permission,
monitoring, and submission stages.
Follow-up
Who can authorise additional time, a restart, or rescheduling?
13
Evidence review
What evidence must a reviewer inspect before taking action?
Review relevant recording segments, technical events, candidate messages, proctor notes, session history, policy rules, assessment context, and limitations.
Listen for
A documented evidence standard rather than reliance on a single
score or alert.
Follow-up
How are confirmed facts separated from assumptions in the review
record?
14
Reviewer quality
How are proctors and reviewers trained and calibrated?
Examine policy training, sample cases, independent review, disagreement resolution, escalation, quality checks, and recurring calibration.
Listen for
Evidence that reviewers can apply the same criteria consistently
across similar sessions.
Follow-up
Which cases are double-reviewed to measure agreement?
15
Human oversight
Can an automated signal directly invalidate or fail a candidate?
Determine whether technology can trigger automatic outcomes and where human review, challenge, correction, and approval are mandatory.
Listen for
Accountable human decision-making for consequential outcomes and
uncertain evidence.
Follow-up
Who has authority to reverse an automated or proctoring outcome?
16
Privacy
How will candidate recordings and session data be protected?
Review access, encryption, exports, downloads, integrations, sharing, reviewer permissions, storage location, audit logs, retention, and deletion.
Listen for
Data controls connected to purpose, role-based access, and a
documented retention schedule.
Follow-up
What happens to downloaded or exported proctoring evidence?
17
Fairness
How will candidate-group differences be monitored?
Compare access, verification, technical events, alerts, review, escalation, invalidation, completion, appeals, and progression.
Listen for
Appropriate segmentation, sufficient context, sample-size
awareness, investigation ownership, and corrective action.
Follow-up
Could device, network, location, language, or accessibility
differences explain the pattern?
18
Incident management
How are complaints, appeals, errors, and disputed conclusions handled?
Define reporting, evidence preservation, investigation, response times, correction, communication, decision authority, and case closure.
Listen for
A visible process that allows relevant evidence to be corrected
or reconsidered.
Follow-up
Is the original reviewer excluded from the appeal when
appropriate?
19
Operational capacity
Can the organisation support the expected review workload?
Estimate session volume, alert rates, review time, peak periods, escalation demand, language coverage, support demand, and service expectations.
Listen for
Staffing and review capacity tested against realistic assessment
volume rather than ideal assumptions.
Follow-up
What happens when the review queue exceeds capacity?
20
Post-launch review
How will the proctoring process be reassessed after implementation?
Review candidate completion, technical incidents, support, accessibility, alert quality, reviewer agreement, escalations, complaints, fairness, and final outcomes.
Listen for
A recurring review schedule with owners, thresholds, evidence,
approvals, and corrective-action tracking.
Follow-up
Which changes require a new pilot or formal reapproval?
Model-specific questions
Ask different questions for different proctoring models
The operating model changes who observes the candidate, when evidence is reviewed, how quickly support can respond, and how much automated interpretation influences the process.
LP
Live proctoring
Questions about real-time supervision
Confirm proctor capacity, communication, intervention, escalation, training, evidence notes, and candidate support.
How many candidates can one proctor supervise responsibly?
Which events allow a proctor to interrupt the assessment?
How are proctor instructions and candidate responses recorded?
RR
Record and review
Questions about delayed evidence review
Review recording quality, queue volume, reviewer access, turnaround, retention, candidate communication, and escalation.
Which sessions enter review and who prioritises the queue?
How long does a complete evidence review normally require?
When is the candidate informed that review is pending?
AP
Automated proctoring
Questions about signals, thresholds, and limitations
Examine signal definitions, false alerts, data inputs, sensitivity, supported environments, human review, and model changes.
Which event types generate an automated alert?
How are thresholds validated and updated?
Can a candidate outcome occur without human review?
HP
Hybrid proctoring
Questions about the boundary between technology and people
Clarify how alerts reach proctors, when live intervention occurs, what requires later review, and who makes the final decision.
Which signals are shown to the live proctor?
Which cases require independent post-session review?
How are conflicting automated and human findings resolved?
Candidate perspective
Can candidates answer their own questions about the process?
Candidate-facing information should make the proctored journey understandable before personal information, recordings, or assessment responses are submitted.
Why is this assessment proctored?
Explain the assessment purpose and how proctoring supports the process.
What will be recorded or monitored?
Identify the camera, microphone, screen, browser, identity, and session information involved.
What should I do when something fails?
Provide preparation help, live support, retry rules, and incident communication.
Who sees my information and how is it used?
Explain access, review, retention, deletion, decision use, and relevant challenge processes.
Evidence-quality questions
Test whether the answers are supported by usable evidence
A confident answer is not enough. Ask how the process is documented, tested, reviewed, monitored, corrected, and connected to candidate and assessment outcomes.
Design evidence
Documents supporting the planned process
Confirm that the proctoring purpose, configuration, candidate journey, evidence rules, and decision responsibilities exist outside the platform interface.
Is there an approved proctoring specification?
Are monitoring controls mapped to identified risks?
Are prohibited uses and decision limits documented?
Is every question assigned to an accountable owner?
Testing evidence
Results from realistic candidate and reviewer journeys
Evaluate normal, interrupted, supported, accommodated, expired, retried, appealed, and high-volume assessment scenarios.
Were supported browsers and devices tested?
Were technical interruptions and recovery tested?
Were accommodation journeys tested end to end?
Were reviewers calibrated using sample sessions?
Operational evidence
Information showing what happens after launch
Review performance through candidate access, technical events, support, alert quality, reviewer agreement, escalations, complaints, fairness, and outcomes.
Are alert and dismissal rates reviewed together?
Are technical issues separated from conduct concerns?
Are candidate-group differences investigated?
Are corrective actions tracked to closure?
Illustrative inquiry log
Record questions, evidence, gaps, owners, and decisions
A structured log can prevent important questions from being answered informally and forgotten. The interface, status labels, and findings below are illustrative.
IQ
Online Proctoring Inquiry Log
Illustrative view
Illustrative inquiry cycle
Questions awaiting evidence or corrective action
Illustrative implementation stage
Illustrative question register
Open questions by implementation area
01
Has the alternative identity route been tested?
Open
02
Can reviewers inspect technical events beside alerts?
Review
03
Who approves a candidate retry after an upload failure?
Owner
04
Are exported recordings covered by deletion rules?
Privacy
05
How will reviewer agreement be monitored?
Quality
Illustrative findings
Evidence gaps requiring action
Accessibility
Alternative assessment instructions have not been tested
with the configured browser controls.
Technical evidence
Reviewer reports currently show alerts without complete
network-event context.
Decision ownership
Retry authority is divided between support and assessment
teams without a final escalation owner.
Data governance
Exported evidence requires a separate retention and deletion
procedure.
All interface elements, questions, statuses, findings, and workflow labels are illustrative. Actual evidence requirements, owners, approvals, timeframes, policies, and actions should reflect the assessment context and governance framework.
Ask who can decide, challenge, correct, and approve
Online proctoring can influence access, assessment validity, candidate progression, academic outcomes, recruitment decisions, privacy, and fairness. Assign accountable owners before consequential cases arise.
Who makes the final assessment-integrity decision?
Distinguish proctor observations, reviewer findings, automated signals, assessment-owner decisions, and appeal outcomes.
Can the decision owner inspect all evidence and documented
limitations?
Who can question or correct an incorrect conclusion?
Define candidate complaints, internal challenge, independent review, evidence correction, outcome reversal, and communication.
Is there a documented route for correcting technical or reviewer
error?
Who approves changes to monitoring and decision rules?
Track changes to identity checks, permissions, alerts, models, thresholds, retention, reports, and candidate communication.
Which changes require testing, a pilot, or formal reapproval?
Who reviews candidate impact after launch?
Assign ownership for technical performance, support, accessibility, fairness, privacy, reviewer consistency, complaints, and outcomes.
How are unresolved findings escalated and tracked to closure?
Frequently asked questions
Online Proctoring Questions FAQs
Review common questions about proctoring purpose, models, identity, technology, monitoring, candidate communication, accessibility, privacy, evidence review, fairness, and governance.
What is the first question to ask about online proctoring?
Ask why proctoring is required for the specific assessment. The
answer should identify a documented assessment risk, explain why the
proposed controls are proportionate, and consider whether a less
intrusive alternative could address the same risk.
How should an online proctoring model be selected?
Compare live, recorded, automated, and hybrid models using the
assessment consequence, scale, candidate needs, support
requirements, review capacity, evidence standards, privacy,
accessibility, cost, and expected response time.
Which identity-verification questions should be asked?
Ask what information is collected, why it is necessary, who reviews
it, how long it is retained, how failed checks are resolved, which
alternatives are available, and whether identity evidence is kept
separate from assessment scoring.
What should candidates know before a proctored assessment?
Candidates should understand the proctoring purpose, model,
equipment, browser, permissions, identity steps, room requirements,
allowed resources, recording, monitoring, support, accommodations,
data handling, expected duration, and incident process.
What should be asked about automated proctoring alerts?
Ask which events generate alerts, which data sources are used, how
thresholds are selected, how accuracy is tested, which limitations
exist, how false alerts are monitored, and whether a person reviews
the evidence before any consequential action.
How should technical failures be handled?
Define pause, reconnect, resume, retry, reschedule, lost-progress,
upload, submission, support, evidence preservation, candidate
communication, and escalation rules. Technical events should not be
treated automatically as candidate misconduct.
Which accessibility questions should be included?
Review camera and microphone requirements, time limits, identity
checks, screen monitoring, browser restrictions, assistive
technology, breaks, room requirements, confidential accommodation
requests, alternative arrangements, support, and successful
completion.
What should be asked about proctoring evidence review?
Ask which evidence reviewers inspect, how facts are separated from
assumptions, how technical context and candidate explanations are
considered, how reviewers are trained, when cases are escalated,
and who owns the final decision.
How should privacy and retention be evaluated?
Review collection purpose, permissions, access, recordings,
transcripts, screen captures, identity data, session logs, alerts,
proctor notes, exports, integrations, storage, retention, deletion,
incident handling, and audit records.
How can CloudTest support online proctoring workflows?
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.
Ask before monitoring, recording, or deciding
Build online proctoring around clear answers and accountable evidence
Define the purpose, select the appropriate model, support candidate access, test the technology, limit monitoring, protect data, interpret alerts carefully, require human review, monitor fairness, and document who owns each decision.