Back to all articles

An exam is supposed to measure what a student knows. When the exam itself is hard to use — for a student who navigates by screen reader, who needs extra time, or who cannot sit still under a webcam without being flagged — it stops measuring knowledge and starts measuring the student's ability to cope with a badly designed tool. Accessibility is the discipline of removing those irrelevant obstacles, and for online assessment it is both a legal duty and a matter of basic fairness.

This guide covers the essentials: what the law expects, what "accessible" means in practice, and — a point specific to remote assessment — how proctoring can be made to accommodate disabled students rather than penalise them.

The legal baseline

The specifics vary by country, but the underlying obligation is remarkably consistent. In the United States, the Americans with Disabilities Act (ADA) and Section 504 of the Rehabilitation Act require that students with disabilities receive reasonable accommodations and that programmes be accessible. In the United Kingdom, the Equality Act 2010 imposes a duty to make "reasonable adjustments." Across the EU and beyond, similar duties apply, and public-sector bodies are increasingly required to meet defined technical accessibility standards.

The common thread: you cannot lawfully run an assessment that a disabled student is unable to access on an equal footing, and "we didn't think about it" is not a defence. Accessibility has to be designed in, not retrofitted after a complaint.

WCAG: the technical standard, in brief

The Web Content Accessibility Guidelines (WCAG), published by the W3C, are the de facto standard for digital accessibility, and any browser-based exam should be built against them. WCAG is organised around four principles, usually remembered by the acronym POUR:

You do not need to memorise the specification to make good decisions. If your exam works fully with a keyboard, reads correctly in a screen reader, has readable contrast, and does not impose a rigid countdown that cannot be extended, you have cleared the most common and most serious barriers.

The accommodations that matter most

Extended time

The single most common accommodation is additional time — often 25% or 50% extra, sometimes more — for students with conditions affecting reading speed, processing, or motor control. An online exam platform must be able to grant extra time to specific individuals without announcing it to the class and without the timer or the auto-submit fighting the accommodation. If extended time has to be arranged by emailing a proctor mid-exam, the system has failed.

Assistive technology compatibility

Screen readers, screen magnifiers, speech-to-text, and alternative input devices must be able to operate during the exam. This is where lockdown approaches often collide with accessibility: software that aggressively blocks other applications can also block the assistive technology a student depends on. The resolution is not to abandon security but to build in explicit allowances for approved assistive tools.

Breaks and flexible scheduling

Some students need rest breaks — for medical reasons, for medication, or to manage fatigue or anxiety. A well-designed exam can pause the timer for a scheduled break without treating the pause as suspicious behaviour.

Alternative formats

Questions that rely on colour alone, that embed text in images a screen reader cannot read, or that require precise mouse dragging can each exclude a group of students. Providing the same assessment in a format that does not depend on a single sensory or motor channel is often straightforward once you are looking for the problem.

The proctoring blind spot

Here is the issue that gets remote assessment into trouble most often. Automated proctoring systems flag "anomalies" — looking away from the screen, moving unusually, another voice in the room, an inconsistent face, long pauses. For a disabled student, many of these are not anomalies at all. A student with a motor or visual disability may not maintain steady eye contact with a camera. A student who uses speech-to-text will produce voice in the room. A student with a support worker present has another person nearby by necessity. A student with a facial difference may confuse a naive face-matching check.

The design rule: automated flags must feed human review, never automatic penalties — and reviewers must know which students have accommodations so that expected behaviour is not misread as misconduct. A system that auto-terminates an exam on a "violation" will, sooner or later, terminate a disabled student for doing nothing wrong.

This is one more reason the industry consensus has moved toward treating proctoring signals as prompts for a human to look, not as verdicts. The move protects everyone, but it protects disabled students most, because they are the group most likely to be flagged for reasons that have nothing to do with integrity.

A practical checklist

  1. Can every part of the exam be completed with a keyboard alone?
  2. Does it read correctly in a screen reader, with real text rather than text baked into images?
  3. Is colour contrast sufficient, and is no information conveyed by colour alone?
  4. Can extended time and rest breaks be granted per student, quietly and reliably?
  5. Does any lockdown or monitoring allow approved assistive technology to run?
  6. Do proctoring flags go to a human who knows who has accommodations — never to an automatic penalty?
  7. Have you asked disabled students themselves what breaks down? They are the best testers you have.

Accessibility is sometimes treated as a compliance burden bolted on at the end. It is more useful to see it as a design constraint that makes the assessment better for everyone: clearer instructions, readable layouts, flexible timing, and human judgement over automated accusation are improvements that no student has ever complained about.

References & Further Reading