1,943 Papers Searched, 14 Made the Cut: Are We Measuring What Helps Inclusive Apprenticeships?

A large search and a surprisingly small evidence shelf

Imagine an apprentice who receives three sensible changes: instructions in advance, a quieter place for concentrated work and the option to explain an early idea before writing it up. The next submission is stronger. The tutor is encouraged; the employer is pleased; the learner feels that the programme has finally become navigable. But what, exactly, improved? Was it the quieter setting, the clearer sequence, growing familiarity with the task, better feedback, more time in the role or some combination of them? This is an illustrative situation, not a Kent Business College learner record. It reveals a mature question for inclusive education: not whether support is kind, but whether a particular change gives a particular learner a fairer route to ambitious occupational performance.

A 2025 systematic review of neurodiversity in post-secondary computing education searched 1,943 records across ACM Digital Library, IEEE Xplore and Scopus. After its defined screening and eligibility process, only 14 papers met the review’s narrow scope; eight contained user studies, and only four of those shared study materials. The ratio is not a verdict on all neurodiversity research, all education or all apprenticeships. It reflects one review’s databases, search terms and focus on post-secondary computing education. Yet it is provocative for digital, data and technology apprenticeships—and instructive for the wider sector—because many curricular recommendations in that evidence set were not underpinned by empirical evaluation.^[5]

The right response is neither cynicism nor paralysis. A provider does not need a randomised trial before giving a learner an accessible document or honouring a reasonable adjustment. Legal duties, ethical practice, professional judgement and the learner’s own account matter. Nor should local monitoring be used to make the apprentice repeatedly prove a disability or to withdraw support at the first uneven week. The opportunity is more constructive: agree what barrier the adjustment is intended to reduce, deliver it reliably, examine relevant learning evidence and improve it with the learner. Inclusion becomes a disciplined partnership rather than an attractive statement.

England’s evidence problem is practical, not theoretical

The latest official context makes this question timely. Department for Education assurance findings, updated on 30 September 2026, report that most learning-support overclaims identified in the 2024 to 2025 assurance work arose where there was insufficient evidence that a reasonable adjustment had been delivered or was in place for each month claimed. Other issues included unclear plans, unclear delivery records and unclear reviews. The publication does not quantify the number of providers or claims behind that phrase, and it is a compliance report rather than an evaluation of teaching effectiveness. It therefore cannot tell us that poorly documented support was educationally ineffective. It does show that good intentions, funding records and dependable delivery can come apart.^[1]

Current England guidance makes the expected sequence unusually clear. For a full apprenticeship, eligible learning support is a fixed £150 for each month in which a costed reasonable adjustment is needed, delivered and evidenced; costs above that rate follow separate excess-support arrangements. The first review should occur one month after implementation, with continuing need reviewed at least every three months. The apprentice must agree arrangements in writing, and evidence should confirm whether adjustments are effective and how they have been adapted. The £150 is not a value placed on inclusion, an automatic payment for a diagnosis or a budget that every useful adjustment must consume. Workplace adjustments funded or provided by the employer are also distinct from provider learning support.^[2]

Ofsted’s monitoring guide for use from September 2026 strengthens the quality argument. For new apprenticeship providers, the inclusion theme asks how much progress leaders have made in identifying individual needs and putting suitable support in place, particularly for apprentices with SEND, disadvantage or other barriers. The guide also asks leaders to demonstrate the positive impact of their work. That is not an instruction to manufacture an effect size, nor does Ofsted prescribe the method proposed in this article. It is a reminder that activity and impact are different claims. A list of adjustments answers “what did we arrange?”; a quality conversation must also ask “was it delivered, what did the learner experience, and what does the work now show?”^[3]

Student voice is evidence—but it is not the whole evidence

The newest research gives learner voice a central place while warning against easy generalisation. Nieminen’s study, published online in October 2025 and appearing in a 2026 volume of Higher Education Research & Development, analysed qualitative survey responses from 139 students with disabilities and followed these with nine narrative interviews at one Finnish university. Participants imagined assessment that was more relational, authentic and critically purposeful, including meaningful tasks and alternatives to narrow conventions. This was qualitative “utopia as method” research: it generated design principles from lived experience, not causal estimates of attainment, completion or employment. It concerned higher education, not English apprenticeships. Its transferable lesson is that learners can identify construct-irrelevant barriers and suggest better ways to make competence visible.^[6]

A broader 2024 meta-ethnographic review synthesised 42 qualitative studies involving 868 disabled students in higher education. It found that assessment could exclude through immediate access barriers—such as inaccessible places or systems—and through a deeper restriction of participation and belonging. Because the synthesis was qualitative, it should not be converted into a percentage of apprentices who will benefit from flexible assessment. It is valuable because it distinguishes access to the assessment from membership in the learning community. An apprentice may receive extra time yet still be excluded from the conversations, workplace practice and feedback through which occupational judgement develops.^[7]

A 2023 systematic scoping review widened the lens further. It included 133 studies covering 12,202 participants across qualitative, quantitative and mixed-methods research on disability, medical conditions and mental health in tertiary education. Students valued individual accommodations, but the synthesis also called for transformation of physical, social and teaching environments, better information and more effective support services. The studies were heterogeneous, and tertiary participation is not apprenticeship achievement. Still, the pattern challenges a false choice between universal design and individual adjustment. Accessible baseline teaching can remove common barriers; an individual agreement can address needs that remain. Neither should be treated as a reason to ignore the other.^[9]

The research is promising, modest and uneven

McDowall and Kiseleva’s 2024 rapid review examined studies summarised in existing reviews of support for neurodivergent students in higher education. It considered examination adjustments, explicit and strategy instruction, technology, psychological support, mentoring, coaching, comprehensive programmes and transitions. The authors concluded that the evidence was modest, often based on small United States samples and more likely to measure skills or learning than successful completion or transition into work. A rapid review is deliberately streamlined and inherits the limitations of the reviews and studies it includes. It nevertheless provides a useful warning for apprenticeship leaders: a fashionable practice can be plausible, welcomed and still under-evaluated for the outcome that matters.^[8]

Filippou, Acquah and Bengs’ 2025 systematic review reached a complementary conclusion at institutional level. Six databases yielded 43 studies of inclusive policies and practices in higher education published between 2001 and 2021; 33 focused on students with disabilities, and only nine studies in the European subset were from the UK. The review shows growing attention but an evidence base distributed across different countries, groups and practices. It does not establish that a named policy will cause apprenticeship success. It does suggest that inclusion cannot live only in a specialist support office: policy, curriculum, staff capability, technology and everyday relationships have to connect.^[10]

This is where a careful provider can add value without pretending to conduct publishable research. Local improvement evidence will not settle scientific questions and should not be marketed as proof for every learner. It can answer a more modest question: did we deliver the agreed change, and is there enough evidence to retain, adapt or replace it for this learner and task? That evidence can include the apprentice’s account, a tutor’s observation, completion of a relevant step, the accuracy or independence shown in comparable work, and the employer mentor’s observation of workplace transfer. More data is not automatically better. The aim is a small set that is meaningful, proportionate and safe.

The CLEAR review: five questions, one ambitious task

We propose a practical routine called CLEAR: Clarify, Link, Enable, Assess, Review. It is an original quality-improvement proposal, not a validated instrument or a replacement for statutory duties, specialist advice, funding rules or assessment-organisation decisions. Start by clarifying one barrier in the learner’s own terms. Avoid beginning with a label and guessing the solution. Ask what happens in a specific task: does the apprentice lose the sequence in a rapid verbal briefing, struggle to navigate a dense template, become overwhelmed by simultaneous discussion, or need a different way to plan before demonstrating the same competence? Record only information necessary for support, with appropriate consent and privacy.

Link the proposed adjustment to the task and the intended standard. If the apprentice must analyse project risk, the standard may require accurate judgement and clear communication; it may not require decoding an unnecessarily crowded slide at speed. If spoken interaction is itself an occupational requirement, do not remove it without considering the standard. Instead, teach and practise it through accessible preparation, an agreed structure and progressively realistic conditions. This is how inclusion and ambition stay together: remove a barrier that is irrelevant to the competence, then preserve the challenge that defines the competence.

Enable means delivering the arrangement consistently across the places where learning happens. A transcript available in college but not for an employer briefing is only a partial pathway. A mentor who has not seen the agreed approach may unintentionally recreate the barrier. Decide who will do what, when, with which accessible materials, and how the apprentice can say that it was missing without fear. Universal features—clear headings, readable documents, predictable navigation, worked examples and advance purpose—can support many learners. Individual adjustments remain individual: a choice of response format may help one person and create a new difficulty for another.

Assess one or two relevant indicators before and after the learner has had a fair opportunity to use the change. Choose evidence close to the task: accuracy of a calculation, completeness of a handover, independence in selecting evidence, quality of reasoning, successful use of a feedback point, or the number and type of prompts needed. Add the apprentice’s judgement: what became easier, what stayed difficult and what unintended burden appeared? Do not use attendance or satisfaction alone as proof of learning. Do not compare one learner casually with another whose role, experience and support differ. An improvement after the adjustment is encouraging, but without a counterfactual it does not establish sole causation.

Review at the agreed point. Keep what is useful; adapt what is partly useful; stop what is unnecessary or burdensome only through a proper conversation and within applicable requirements. If performance has not improved, the conclusion is not “the learner failed the adjustment”. The task may be too broad, delivery inconsistent, the indicator poorly chosen, the learner under-practised, or the proposed change simply wrong. Where the barrier or decision is specialist, seek qualified advice. A short honest review is more valuable than a perfect-looking record that tells nobody what happened.

Four roles, and the evidence each one can contribute

For the apprentice, the role is not to disclose everything or design the entire solution. It is to describe the task experience as far as they are comfortable, help judge whether the change is usable and bring examples of work to the review. The tutor translates the occupational goal into teachable steps, checks accessible materials, looks for relevant learning evidence and avoids confusing speed, polish or eye contact with understanding unless the standard genuinely requires them. The workplace mentor creates opportunities to apply the learning, keeps instructions and feedback coherent, and reports observations rather than diagnoses. The quality leader checks whether agreements are timely, delivered and reviewed, and whether patterns point to a curriculum or system barrier.

At cohort level, leaders should use protected, appropriately aggregated information and interpret small numbers cautiously. Useful questions include: how quickly are agreed arrangements implemented; are reviews happening after real use; do some programmes repeatedly produce the same access problem; are learners offered opportunities to practise before high-stakes assessment; and do employer mentors know their part? Outcomes should be disaggregated only where lawful, ethical and statistically sensible. A difference between groups can prompt investigation but does not explain its cause. Administrative learning-difficulty/disability categories are not interchangeable with neurodivergence, and a recorded declaration is not a diagnosis or a complete account of need.

Acas guidance reinforces the individual principle in the workplace: people with the same condition can have different strengths and difficulties, and support may need to be considered without a diagnosis. Possible adjustments include clearer organisation, advance warning of change and changes to the sensory environment, but appropriateness depends on circumstances. The apprenticeship provider, employer and assessment organisation may have different responsibilities. CLEAR should help those parties coordinate evidence; it must not blur who has authority to agree a workplace or assessment adjustment.^[11]

Creativity needs dependable access and room to test an idea

Measurement can sound hostile to creativity when it becomes a narrow scorecard. Used well, it can do the opposite. An apprentice is more likely to propose an original improvement when the problem is understandable, the route to participation is accessible and early ideas can be examined without humiliation. Give the learner a real occupational problem, time to explore more than one response and a clear account of the constraints. Let them sketch, model, annotate, discuss or prototype where those routes do not change the competence being assessed. Then ask for evidence, comparison and refinement. Creativity is not a diagnostic “superpower” and should not be demanded as compensation for difficulty. It is a practice that can be taught, supported and judged.

The strongest local evidence may therefore be a sequence rather than a score: the apprentice identifies a barrier; an agreed adjustment creates access; practice reveals a misconception; feedback improves the reasoning; and the learner transfers that reasoning to a new problem. That story preserves challenge and shows development. It also produces better management information than “support provided”. If the creative contribution improves a process, record the change separately from the learning claim. The apprentice’s idea may be valuable even when the evaluation cannot prove that an adjustment caused it.

The research gap identified in computing education should make providers curious, not timid. Apprenticeships are unusually well placed to connect teaching with live occupational practice. With consent and good governance, providers can learn systematically from what learners, tutors and employers observe. They can share methods transparently, distinguish internal improvement from generalisable research and collaborate with researchers on stronger studies. The goal is not to turn every review meeting into a laboratory. It is to stop losing useful knowledge because nobody agreed what to look for.

A Monday promise: support that learns

Return to the apprentice whose work improved after three changes. A weak system files the success as proof that the package works. A punitive system waits for a mistake and removes it. A learning system asks the apprentice which element changed access, checks whether it was delivered across settings, examines the next comparable task and refines the agreement without lowering the occupational standard. It can celebrate progress while remaining honest about uncertainty. That is evidence-informed inclusion in its most practical form.

The hopeful conclusion is that a thin international evidence base does not leave educators empty-handed. It tells us to listen carefully, design access in advance, individualise where necessary, measure proportionately and share what we learn with humility. England’s funding and inspection frameworks already emphasise assessment, delivery, review and impact. The opportunity is to make those processes educationally useful—not merely compliant—so that the record helps the next tutor, the next mentor and, above all, the apprentice.

This week, choose one ambitious task and try the CLEAR questions with the learner: what is the barrier, how does the proposed change connect to the standard, was it delivered, what evidence matters, and what should we keep or change? Our discussion question is: what is one sign—beyond attendance or satisfaction—that tells you an adjustment is helping a learner become more capable? Please share a practical indicator without disclosing anyone’s diagnosis or private circumstances, and pass the article to a tutor, mentor, employer or quality leader who could test it. New editions of The Monday Common Room are released every Monday at 9:00 am UK time; readers can follow the RSS feed or add the publication calendar below.