top of page
Search

Psychometric Tests & Online Assessments (UK): Reasonable Adjustments That Hold Up


Hiring should measure what matters for the job—not how fast someone can decode tiny text in a noisy room or navigate an inaccessible testing platform. This guide sets out lawful, evidence-based adjustments for psychometric and online assessments in the UK, so you can keep validity high and remove avoidable barriers.


Your legal footing (in plain English)

Under the Equality Act 2010, employers must make reasonable adjustments in recruitment where a disabled candidate would otherwise be placed at a substantial disadvantage. That duty applies to tests and online assessments, not just interviews. The EHRC’s Employment Statutory Code of Practice makes this explicit, and GOV.UK confirms you can (and should) invite candidates to request adjustments during recruitment.


What a “valid” adjustment looks like

Good adjustments do not change the construct you’re measuring; they change access to the test. The British Psychological Society (BPS) advises test users to plan for adjustments, train administrators, and ensure that candidates with disabilities can complete assessments fairly. The International Test Commission (ITC) goes further: if a timed multiple-choice test is really measuring reading speed, consider an alternative assessment that measures the target skill directly (for example, a scored work sample). Document the rationale either way.


Digital accessibility is not optional

If you use online assessments, your platform should meet WCAG 2.2 AA at minimum and be usable with screen readers, magnifiers and keyboard navigation. UK government service guidance is blunt: if a digital service does not meet WCAG 2.2 AA, you may be breaking the law. Test critical journeys with assistive tech in the stack (for example, JAWS/NVDA with mainstream browsers), as RNIB recommends in its audits.


Practical adjustments that preserve fairness

Formats and delivery. Provide larger text, clear line spacing, and dyslexia-friendly layouts; allow use of personal assistive tech (text-to-speech, screen readers, dictation) where it does not alter the construct. Offer on-screen or paper options when relevant, and allow a quiet room or noise-reducing headsets for in-person tests. The BPS notes that policy should explicitly cover adjustments and that test users must be trained to implement them.


Timing and pacing. Extra time can be appropriate when a disability affects reading rate or processing speed but speed is not the skill you’re hiring for. Where the construct is intrinsically speeded (e.g., live data entry under time constraints), use a job-relevant work sample that demonstrates that specific speeded competence rather than a generic timed puzzle. ITC guidance explicitly encourages considering alternative procedures when modifications would compromise validity.


Breaks and chunking. Allow short, scheduled breaks to manage attention, pain, or sensory overload, with test security maintained. Provide practice items and plain-English instructions so working memory isn’t the bottleneck. These steps improve measurement for all candidates without lowering standards.


Remote and proctored tests. Publish the set-up (lighting, camera, room rules) well in advance and state how candidates can use assistive tech under remote proctoring. If the proctoring client blocks screen readers or dictation, you must supply an equivalent route (for example, an alternative browser session, an on-site slot, or a different assessment of the same construct). Align your vendor and policy to WCAG 2.2 AA.



When to switch to a work sample

If the online test confounds the construct with barriers you can’t safely remove (for example, a game-based assessment that is not accessible to screen readers), replace it with a scored work sample mapped to the role’s competencies. The ITC guidelines explicitly recommend alternative structured forms of assessment in such cases. Record the mapping and keep the scoring rubric.


Process that stands up to scrutiny

Write a short adjustments policy for selection that covers: how candidates tell you what they need; who decides; how decisions are recorded; and how you assure accessibility of third-party platforms. ACAS’s current guidance on reasonable adjustments supports a needs-led approach without gatekeeping—diagnosis is not a prerequisite to begin proportionate changes. The policy should also name a review point so practices improve over time.


For candidates: a concise request that works

You don’t need clinical detail. Explain the barrier and the change that removes it.

Subject: Adjustment for online assessment

 Hello [Recruiter], I’m applying for [role]. Because of [brief descriptor, e.g., a disability affecting reading rate], the default set-up would disadvantage me. I do this fairly with [example: 25% extra time, ability to use text-to-speech, and a quiet space]. These changes don’t alter what you’re measuring. Please confirm how I can enable them on your platform, or suggest an equivalent route if the platform can’t support them. Many thanks, [Name].

GOV.UK confirms employers should make reasonable adjustments during recruitment; reference that if needed.


Common pitfalls (and how to avoid them)

Assuming “extra time” fixes everything. Sometimes layout, platform incompatibility, or sensory load is the real problem; fix the format first, then consider time.


Buying inaccessible tests. If your vendor cannot evidence WCAG 2.2 AA and real assistive-tech testing, don’t deploy their product. Demand an accessibility statement and a test plan. 

Measuring the wrong thing. If success in your test relies on puzzle-solving with unfamiliar rules, ask whether that skill truly predicts performance in the role. Prefer validated work samples or structured tasks tied to the job.


Poor records. Keep a simple note of what was requested, what you granted, and why. That shows you considered reasonableness and protects both candidate and employer under the Equality Act framework.


References (APA-7)

Comments

bottom of page