A record was eligible for inclusion if it met all five of the following criteria:
- IC1. The study addresses the intersection of neurodivergence — including autism spectrum condition, ADHD, dyslexia, dyspraxia, dyscalculia, Tourette syndrome, or broader neurodiversity — AND work, employment or labour market participation.
- IC2. The population includes adults aged 16 or over, or the study addresses transition from education to employment.
- IC3. The study is empirical (quantitative, qualitative or mixed methods), theoretical, conceptual, or a literature review (including systematic reviews, scoping reviews, and mapping reviews).
- IC4. The publication is a peer-reviewed article, book chapter, preprint with sufficient methodological transparency, or grey literature produced by a recognised academic or policy institution.
- IC5. Full text is retrievable in English, Spanish, or Russian.
A record was excluded if it met any of the following criteria:
- EC1. The study addresses neurodivergence exclusively in clinical or diagnostic contexts without reference to occupational, social or economic function.
- EC2. The population is exclusively children or adolescents with no reference to transition to adulthood or employment.
- EC3. The study addresses only assistive technology or medical intervention without reference to workplace or labour market outcomes.
- EC4. The record is a conference abstract, editorial, letter or book review without substantive empirical or theoretical content.
- EC5. Full text is not retrievable.
Six primary databases were searched with no start-date restriction:
| Database | Search period | Records (n) | Notes |
|-----------|---------------|-------------|-------|
| PubMed | 2025–2026 | 90 | Peer-reviewed biomedical and social science |
| OpenAlex | 2009–2026 | 200 | Open academic graph; broadest disciplinary coverage |
| ERIH PLUS | 2007–2026 | 502 | European humanities and social sciences index |
| ERIC | 2020–2024 | 3 | Education research database |
| medRxiv | 2009–2026 | 70 | Health sciences preprints |
| bioRxiv | 2009–2026 | 41 | Life sciences preprints |
**TOTAL 906 Before deduplication
Complementary searches were conducted in Russian (CyberLeninka, 2022–2024; search terms: дискриминация Трудоустройством аутизм; n=60 retrieved, n=6 selected) and Spanish (Google Scholar, 2024–2026; search terms: neurodiversidad y trabajo; n=39 retrieved, n=39 selected). Complementary records do not enter the primary corpus bibliometric counts but are integrated analytically in the narrative synthesis with explicit attribution.
Search terms combined three conceptual domains across all six databases:
- Domain 1 — Neurodiversity/neurodivergence terminology: neurodiver*, autis*, ADHD, dyslexia, dyspraxia, dyscalculia, Tourette, neurodivergent
- Domain 2 — Work and employment terminology: employ*, work*, labour market, labor market, workplace, job, vocation*, occupation*
- Domain 3 — Organisational terminology: organisation*, organization*, manage*, HRM, human resource*, inclusion, diversity, accommodation
Boolean operator: AND between domains; OR within domains. No start-date restriction. No language filter applied in primary search. Full search strings for all six databases are available from the corresponding author on request.
Title and abstract screening was conducted using Abstrackr (abstrackr.com), an online systematic review management platform. Primary screening was conducted by the lead author. A second reviewer independently screened a 10% random subsample (n=47 records); inter-rater reliability was assessed using Cohen's kappa (κ = 0.84), indicating strong agreement.
Records excluded at title and abstract stage were documented with reasons:
- Did not address the intersection of neurodivergence AND work/employment: n=41
- Exclusively clinical diagnosis without reference to occupational function: n=24
- Exclusively child population without transition component: n=12
- Conference abstract without substantive content: n=6
Records proceeding to inclusion: n=383 (primary corpus).
2.6 Data Charting Process
Data were charted by the lead author using a standardised extraction form developed a priori. A second reviewer independently charted a 15% random subsample for reliability assessment of theoretical orientation coding (κ = 0.79). Discrepancies were resolved through discussion.
- Declared geographic context and country of study (missing coded as 'not specified')
- Journal title and disciplinary classification
- Methodological design — 5 categories: qualitative; quantitative; mixed methods; systematic/scoping review; theoretical/conceptual
- Theoretical and disciplinary orientation — 7 inductively coded categories: (1) diversity management/HRM paradigm; (2) neurodiversity paradigm; (3) individual psychological/cognitive framework; (4) identity/stigma/disclosure; (5) intersectionality/gender; (6) critical political economy; (7) critical disability studies
The analysis proceeded in two phases. Phase I comprised descriptive bibliometric mapping: frequency analysis and proportional distributions for annual publication counts, geographic concentration, methodological composition, theoretical orientation, and journal concentration. Phase II comprised five-cluster inductive narrative synthesis, refined through two rounds of iterative coding, followed by a cross-cutting analysis identifying the political-economic articulation gap as a structural property of the field. A Bourdieusian structural analysis of field constitution was conducted as a third analytical layer.
2.8 Researcher Positionality
The lead author is autistic. This positionality situates the review within the insider standpoint tradition in disability studies (Oliver, 1992; Barnes, 2003) and satisfies Disability 26 Society's requirement that research be linked to disabled people's interests. The political-economic articulation gap was identified inductively through systematic analysis of corpus metadata; it was not a prior ideological commitment. Experience of the phenomenon constitutes an epistemological resource, not a confound.
3. Anticipated Limitations
The following limitations were identified prior to data collection and are documented here for transparency:
- Anglophone concentration: The primary search databases are predominantly Anglophone. Complementary multilingual searches in Russian and Spanish partially address this but do not eliminate it.
- Single-reviewer charting: Data charting was conducted primarily by the lead author. Inter-rater reliability was assessed on a 15% subsample but full independent dual coding of all 383 records was not feasible.
- Title/abstract coding: Theoretical orientation was coded from titles and abstracts rather than full texts. This may underestimate the presence of critical frameworks in studies whose methodology sections engage more deeply with structural analysis than their abstracts suggest.
- Mapping design: The mapping design does not permit quality appraisal of individual studies or synthesis of effect sizes. Findings describe the structure of the field, not the effectiveness of any specific intervention.
- Complementary searches: The complementary multilingual literature (n=45) was identified through non-systematic manual searches. It is analytically illustrative, not exhaustive.
- No prospective registration: The review was not prospectively registered. This retrospective protocol documents what was done.
Appendix: PRISMA-ScR Compliance Statement
This review was conducted in accordance with the PRISMA Extension for Scoping Reviews (PRISMA-ScR; Tricco et al., 2018) and PRISMA 2020 (Page et al., 2021). All 22 PRISMA-ScR checklist items have been addressed in the manuscript, with items not applicable to a systematic mapping review explicitly noted and justified. A completed PRISMA-COSMIN for OMIs 2024 full-report checklist (Elsman et al., 2024) is available as a separate supplementary document.
PRISMA selection numbers: 906 records identified → 440 duplicates removed → 466 screened → 83 excluded at title/abstract (41 no intersection with work; 24 exclusively clinical; 12 exclusively child population; 6 conference abstracts) → 383 included in primary corpus → complementary searches: 99 identified → 45 included → 428 total corpus for synthesis.**
_End of Protocol Document_
_Version 1.0 · March 2026 · For OSF/Zenodo deposit and manuscript supplementary material_