Online examination security becomes more effective when assessment leaders can explain what each safeguard establishes and how those safeguards work together. Browser restrictions protect a defined part of the digital environment, while identity checks, supervision, assessment design and recovery procedures provide different kinds of assurance. Together, they help universities and vocational education providers maintain conditions appropriate to the consequences of an examination.

A locked browser may prevent candidates from switching applications or opening unauthorised material on the examination device. It cannot independently verify candidate identity or account for activity beyond that device. The useful question is therefore not how many restrictions an institution can introduce, but which measures produce dependable evidence that an assessment was completed under its intended conditions.

Browser lockdown has a valuable and specific purpose. Depending on its configuration, it can limit access to websites, applications, files and copying functions during an examination. Assessment teams can make that protection more reliable by defining permitted resources before the test and checking that the configuration matches those rules.

University College London’s guidance on lockdown browser examinations illustrates why the delivery setting matters. UCL specifies in person invigilation for its lockdown browser examinations and recommends testing equipment, preparing candidates and establishing arrangements for technical support. The browser restriction is one part of a planned examination environment rather than a complete supervision method.

That approach translates into practical preparation. Academic teams can agree which materials candidates may consult, ensure instructions match the available digital tools and arrange a practice examination using the intended device wherever feasible. Administrators can then test settings against those expectations before the live session. These steps make authorised activity clearer to candidates and give invigilators a consistent basis for interpreting incidents.

Device control and candidate authentication answer different questions. One concerns what the examination device permits; the other concerns who is responsible for the submitted work. Assessment leaders can address both without applying identical checks to every examination.

In a supervised campus setting, checking institutional identification against attendance records may provide an appropriate starting point. Remote delivery may justify additional authentication suited to the examination’s stakes and the institution’s procedures. Teams can also specify what happens after an interruption or a change of workstation, when identity may need to be confirmed again.

A useful identity protocol records when verification occurs, who is authorised to perform it and how an exception is escalated. Clear responsibility reduces uncertainty for examination administrators while supporting confidence that an academic decision relates to the authorised candidate’s work.

Invigilation becomes more useful when observations are connected to an agreed review process. Assessment leaders can distinguish three activities: identifying an event, preserving a factual record and deciding whether it merits academic consideration.

A candidate’s unexpected disconnection, for example, may be an ordinary technical interruption. An invigilator can record the time, the system state and the support provided without treating the event as evidence of misconduct. A designated reviewer can then consider that record against the examination rules and other available information.

The same principle applies to suspected unauthorised assistance. Briefing invigilators on what to document and when to escalate supports consistent treatment across rooms, sessions and delivery modes. Academic staff retain responsibility for interpreting evidence and determining a proportionate response. The benefit is not simply closer observation, but clearer, more defensible decisions informed by a reliable record.

Security arrangements become more proportionate when educators consider the demands of the assessment itself. Australia’s Tertiary Education Quality and Standards Agency discusses differing vulnerabilities across assessment formats in its 2026 assessment security guidance on understanding risks. Its toolkit connects the security of an assessment with its format, intended conditions and consequences.

This gives academic teams a constructive starting point: identify the knowledge or competence that must be demonstrated, then select question formats and delivery conditions that produce credible evidence of it. An applied scenario may require candidates to use knowledge in an unfamiliar context, while a carefully constructed bank of equivalent questions can support variation between examination sittings. Neither approach replaces supervision, but each can address risks through the design of the task.

Time limits deserve similar judgement. Restricting the available time more severely does not automatically strengthen validity. Educators can instead establish the time reasonably needed for the intended reasoning, calculation or written response, then decide which materials, tools and accessibility arrangements are compatible with that purpose.

Jisc’s 2025 report, Trends in Assessment in Higher Education: Considerations for Policy and Practice identifies interest in assessment redesign at programme level, blended delivery and reconsideration of traditional invigilated examinations. Such developments reinforce the value of reviewing security requirements alongside assessment purpose rather than treating technical restrictions as a separate decision.

Recovery Planning Preserves Comparable Exam Conditions

Reliable delivery is another form of assurance. A well prepared examination programme establishes in advance how access will be restored, responses preserved and time adjustments considered when a technical interruption occurs.

Assessment teams can rehearse a disrupted session before examination day. The exercise might test whether administrators can identify the last saved response, verify the returning candidate and document any additional time granted. It can also clarify when technical support should resolve an access issue and when academic staff need to decide whether the original assessment conditions remain valid.

Responsibility is easier to coordinate when the process is explicit. Support personnel handle device and connection problems; examination administrators document the interruption and apply authorised procedures; academic decision makers review any implications for fairness or validity.

The arrangements can be assessed through practical indicators such as the proportion of sessions completed without disruption, time taken to restore access and consistency in incident records. These measures offer operational evidence for improving subsequent sittings, rather than assuming that successful access alone proves reliable delivery.

Within this wider planning process, high stake exam software can provide the technological foundation for coordinating examination access, secure delivery and operational incident handling, while assessment leaders retain responsibility for defining procedures and making academic decisions.

A short formative quiz and an examination determining professional competence need not operate under identical conditions. Assessment leaders can begin with the consequences of an unreliable result, identify the most relevant risks and select measures that respond to them.

Where access to unauthorised digital materials is the central concern, browser restrictions may be useful. Where candidate identity is particularly consequential, authentication deserves greater attention. Where an examination depends on live oversight, staff briefings and consistent incident review procedures may add more value than another technical restriction.

The strongest model also considers how controls connect across registration, examination preparation, live delivery and result review. A review of those connections can reveal whether an identity check remains meaningful after reconnection, whether invigilators have the information needed to escalate an incident and whether disruptions are treated consistently between sittings.

Browser lockdown earns its place in that model by doing a defined job well. The wider measure of security is whether educators can explain, with appropriate evidence, why the conditions under which a result was produced justify confidence in the academic judgement that follows.