The Proctoring Question – Part One: Fairness across Delivery Channels
Earlier this year, the Association of Chartered Certified Accountants (ACCA), one of the world’s largest professional bodies with more than half a million students, moved most of its exams back into test centres. ACCA cited the pace of change in AI as a key factor.
The sector is split, and other bodies are watching closely.
For some, that raises the question of whether we risk breaking the promise of fairness and equity that we made in 2020.
In the opening article of this series, we argued that remote proctoring is no longer a product decision but a governance challenge. This first deep dive takes on the theme at the heart of that challenge: when candidates can choose how they are assessed, can the system remain fair?
The rapid expansion of remote proctoring during the pandemic was, at the time, a pragmatic solution to an extraordinary problem. Test centres closed. Travel stopped. Lockdown started. But the show had to go on.
As the pandemic subsided, an appetite for remote assessment remained. It was flexible. Modern. It expanded reach and inclusion. What began as an emergency fix has now evolved into an integral part of assessment delivery.
That shift demands a more rigorous line of questioning. We should no longer be focusing on whether remote assessment works, but on whether a multi-channel system can remain fair, comparable and defensible over time and at scale.
When candidates can choose how they are assessed, flexibility increases.
But so does complexity.
And without deliberate design, complexity becomes risk.
Flexibility is not neutral
Giving candidates the option to sit in-centre or at home feels fair. It feels inclusive. But choice is not inherently fair.
If it is possible for two candidates to achieve the same qualification through materially different assessment conditions, we must ask ourselves whether those conditions produce equivalent challenge, scrutiny and experience.
Fairness cannot be assumed simply because our policies permit multiple routes. It must be demonstrated.
Environment matters more than we admit
We largely control the test centre environment: hardware, seating, lighting, noise levels, identity checks, with invigilators on site to manage them on the day. The candidate’s surroundings are, typically, neutral by design.
A candidate at home brings their own environment with them: a personal device, a personal network, variable connectivity, shared living space, pets, courier deliveries. Their normal daily life is still going on just outside the door.
Most candidates act with integrity in either setting. That is not the point. The point is structural consistency.
In-centre invigilation is visible and predictable, with clear standards of behaviour and someone on hand to help calmly when problems arise. Remote proctoring has become sophisticated, but oversight remains structurally different: dependent on software, bandwidth and camera angles rather than human presence. Where vulnerabilities exist, they are identified, and sometimes exploited. Over time, small advantages compound as workarounds are shared between cohorts.
The risk is not widespread misconduct. It is incremental divergence. And if one route is perceived as more lenient or stricter than the other, we are no longer dealing with a technical design issue; we are dealing with the perception of fairness. Perceptions matter, because they shape trust in both the process and acceptance of outcomes.
What’s at stake
Fairness is not simply about the candidates sitting today. It is about those who have sat in the past, those who will sit in future, and the employers, regulators and wider society who rely on our qualification as a signal of competence and reliability.
If stakeholders begin to believe that the route a candidate selects influences difficulty or scrutiny, the integrity of the qualification is weakened. So we need to ask ourselves whether we can confidently state that a pass achieved remotely is as trustworthy as a pass achieved in a test centre.
If the honest answer is “not yet” or “not always”, then we are morally obliged to act. Not just for today’s candidates, but in fairness to all of those who have previously earned the qualification, and to those who will rely on it in practice.
Choice, access and equity
Remote delivery has expanded access for many candidates, particularly those with accessibility constraints, caring responsibilities or geographical challenges.
That is progress.
But access and fairness are different things.
A candidate with a private room, fibre broadband and high-quality equipment experiences remote assessment differently from someone sharing a crowded flat with an unreliable connection and a decade-old device.
A candidate within easy reach of an assessment centre might prefer the experience of a predictable process, whilst someone in a region with limited test-centre infrastructure may have no realistic option but to sit remotely, even if that means accepting higher levels of stress and technological risk.
In a multi-channel model, structural advantages and disadvantages do not disappear. They are redistributed.
Our obligation, therefore, is not to remove choice, but to ensure that choice does not mask inequality.
What does fairness require?
If multi-channel delivery is here to stay, fairness must be treated as a design discipline, not a soundbite.
If the same behaviour is treated differently depending on location, for example a minor disruption in a test centre is handled informally while a similar disruption at home is escalated as a serious flag, candidates have a valid reason to question fairness.
If we are really committed to delivering fairness, we need to ask ourselves some challenging questions, and answer them frankly:
- Are the rules, expectations and processes clearly and consistently documented for all routes? And do candidates receive equal preparation during formative phases?
- Are the evidential standards and sanctions aligned in a way that ensures a candidate is not treated more harshly simply because they chose one location over another?
- Are our teams suitably trained, and kept up to date with developments, so they know exactly what to look for?
- Are environmental differences and candidate outcome data being formally analysed across routes to evaluate performance impact?
- Are integrity incidents monitored comparatively, not in isolation?
- Do we have mitigation strategies in place where disparities emerge?
These questions are not arguments against remote assessment. Quite the contrary: they are requirements for governing it responsibly.
Demonstrating comparability
Parity of outcomes does not automatically confirm parity of experience.
Comparability must be demonstrated through ongoing monitoring, analytics and review. Where differences appear, even subtle ones, they should be interrogated, not dismissed. Where vulnerabilities emerge, they must be addressed structurally, not reactively.
We must balance oversight with support, and strengthen verification without undermining accessibility.
In a world of rapid change, equivalence is not static. It must be maintained.
Governance over optimism
Too often the remote assessment debate polarises between enthusiastic support and sceptical dismissal. That framing does not help us move forward.
Remote assessment is neither inherently inferior nor automatically equivalent. It is a process, and like any process, its fairness depends on design, implementation and governance.
The strategic task is not to defend flexibility at all cost, but to ensure that flexibility does not compromise standards. Because if standards weaken, inclusion loses meaning.
Good governance is where the concept of fairness becomes reality. Well written, well communicated policies narrow the gap between delivery models. Poorly designed policies widen it, even if the underlying technology is working perfectly.
Fairness by design
If we accept that fairness between in-centre and remote exams is critical at both a candidate and qualification-level, what does that mean for us in practice?
- Start with a fairness statement
The first step is to be explicit about our goal. Fundamentally, a candidate’s result should depend on their ability and performance on assessment day, not on whether they chose to sit in-centre or remotely. This should be at the core of our decision-making and process design. - Align rules and processes across routes
Where possible, we should define misconduct categories, investigation steps and sanctions in a way that they can be applied consistently regardless of route. When there must be differences, these must be communicated transparently. - Surveillance AND Support
Fairness is not only about identifying those who try to cheat, it is also about helping candidates to not fall foul of the guidelines. That means clear pre-assessment guidance, formative assessments which mirror the summative experience, clear technical requirements and responsive, human, support during remote sittings. - Monitor and adjust using data and feedback
Commit to periodic reviews of performance, incident and appeal data across routes. Actively listen to what candidates and staff say about their experiences, and be willing to make adjustments when and where they are necessary. - Communicate our reasoning
Candidates, employers and regulators are more likely to trust our approach if they understand why we do what we do, what constraints we are working with and what activities we undertake to ensure that the integrity of the qualification remains intact.
Trust is our Greatest Asset
Both test centres and remote proctoring are now part of modern assessment delivery. The question is no longer whether one will replace the other, but whether the system that connects them can remain robust and credible over time.
Choice only strengthens assessment when it is underpinned by structural integrity. Without that, flexibility becomes fragility.
Addressing fairness does not require retreat from innovation. It requires discipline. It requires evidence. It requires governance that is as sophisticated as the technology we deploy.
If we get that right, candidate choice becomes a genuine expression of access and inclusion.
If we get it wrong, we erode the most valuable asset assessment has: trust.
Next in the series: the hidden engine room of remote proctoring. We look at reviewer workload and flag fatigue, and why staff wellbeing is an integrity safeguard, not an operational afterthought.
Thinking about what this means for your own assessment model?
Cirrus Assessment works with the assessment sector to evaluate and strengthen assessment delivery, from technical infrastructure to governance frameworks. Talk to our team
