Dictionary · v2026.1 · CC-BY 4.0
The CASRAI Dictionary: A Research Administration Glossary
The CASRAI Dictionary is a citable, machine-readable controlled vocabulary for research administration, containing 1242 definitions across 24 thematic domains in release v2026.1. Stewarded by 20 community working groups; federated with NISO, euroCRIS, CODATA, and RDA; published under CC-BY 4.0 for unrestricted reuse.
What's in v2026.1
- Entries
- 1242
- Domains
- 24
- Tracks
- 5
- Open formats
- 6
- · Stable URIs per entry
- · Schema.org
DefinedTermmarkup - · JATS · JSON-LD · OWL · RDF · CSV · BibTeX
- · Versioned with semver-style release tags
Start here
Browse the dictionary
Five ways into the full corpus, plus the 24 thematic domains below. Every entry is a stable, citable URI under CC-BY 4.0.
/dictionary/browse
Browse by domain
Entries grouped by their thematic domain, with status and CRIS-readiness filters.
/dictionary/a-z
A–Z index of all 1242 terms
Every entry, alphabetised under A–Z headings — the complete map of the dictionary.
/rdm-glossary
Research data management glossary
A curated RDM glossary: data infrastructure, DMPs, persistent identifiers, and reproducibility.
/dictionary/objects
Object templates
123 structured business objects — Award, Book, Conference Paper — for CRIS and RIM integration.
/dictionary/picklists
Controlled picklists
Enumerated value-lists — output types, contributor roles, licence types, access types.
/dictionary/templates
Document templates
23 composite documents — CVs, funding history, DMP ethics review — assembled from object fields.
/dictionary/search
Full-text search
Search definitions, picklists, sample values, and crosswalks with type-ahead suggestions.
Jump to a domain
24 domains across 5 tracks
The full domain map
Each domain is stewarded by a working group. Click any tile to read its entries.
Domain breakdown
Entries by domain · 1242 total
The full distribution of dictionary entries across the 20 stewardship domains, ordered by size. Track colour indicates stewardship cluster.
Compliance and regulatory
Research outputs (expanded)
Funding lifecycle and financial vocabulary
Research integrity and misconduct
Clinical research and clinical trials
Research data infrastructure
Mentorship, training, and career stages
The persistent identifier ecosystem
Research-information systems and integration
Research lifecycle stages and project metadata
Reproducibility and computational research
Grants management and sponsor compliance
CRediT extensions and adjacent contribution vocabularies
Responsible research assessment
Knowledge equity, diversity, global-south inclusion
AI and ML research outputs
Engagement, impact, and SDG alignment
Generative AI use and disclosure
Research security
Machine-actionable data management plans (maDMP)
Sustainable research and laboratory operations
Indigenous data governance — CARE principles
Technology transfer and commercialization
Scholarly publishing
Highlighted terms by domain
One representative entry per domain
A curated tour of high-value definitions — one entry per stewardship domain, chosen to illustrate the breadth of the dictionary in practice.
AI-assisted writing
Text produced with substantive help from a generative AI system, requiring disclosure under most publisher policies.
Lead / equal / supporting qualifier
Modifier appended to a CRediT role to declare relative degree of contribution within that role on a paper.
Narrative CV
Structured-prose researcher biography (Royal Society, UKRI, NIH) replacing publication-list-driven assessment.
Registered report
Two-stage publication where methodology is peer-reviewed and accepted before data collection begins.
ORCID iD
Sixteen-digit researcher identifier resolvable at orcid.org, used to disambiguate authorship across systems.
CERIF
euroCRIS data-model standard for Current Research Information Systems, defining entities for persons, projects, outputs.
Trusted Research Environment
Secure analysis platform where sensitive data is held centrally and researchers bring their code, not the other way around.
Data management plan (DMP)
Living document declaring how research data will be collected, stored, shared, and preserved across the project lifecycle.
Reproducibility
Capacity of an independent party to obtain the same result using the original code, data, and computational environment.
Model card
Structured documentation of an ML model: training data, intended use, performance, limitations, and ethical considerations.
Paper mill
Commercial operation that fabricates manuscripts and authorship slots for sale, a growing integrity threat to journals.
GDPR
EU regulation governing personal-data processing, with extraterritorial reach and direct implications for research data.
NSPM-33
US National Security Presidential Memorandum 33 mandating disclosure of foreign affiliations on federally funded research.
CARE principles
Collective benefit, Authority to control, Responsibility, Ethics — Indigenous Peoples’ data governance principles, complementing FAIR.
DORA
San Francisco Declaration on Research Assessment, calling for an end to journal-impact-factor-based evaluation of researchers.
Plan S
cOAlition S mandate requiring immediate open access for publications arising from publicly funded research.
PPI (patient and public involvement)
Research carried out with or by patients and members of the public, not merely about or for them.
LEAF framework
Laboratory Efficiency Assessment Framework — bronze/silver/gold certification for sustainable laboratory operations.
MTDC
Modified Total Direct Cost — the cost base on which US federal indirect-cost rates are calculated, excluding equipment and subawards above $25k.
Project ID (RAiD-anchored)
Research Activity Identifier (RAiD) handle anchoring a project across funders, institutions, outputs, and contributors.
Featured entries
Selected from v2026.1
A taste of what's inside — definitions, picklists, related terms, structured data, all under CC-BY 4.0.
Kerndatensatz Forschung (KDSF)
The Kerndatensatz Forschung (KDSF) is Germany's national standard defining how universities and non-university research institutions structure and report information about research activity -- projects, personnel, publications, funding, and infrastructure -- using a shared base data model (Basisdatenmodell), topic-scoped modules, and standardised reference queries (Referenzabfragen). Voluntary adoption was recommended by the German Council of Science and Humanities (Wissenschaftsrat) in January 2016; ongoing governance now sits with the Kommission fur Forschungsinformationen in Deutschland (KFiD).
SocArXiv
<p><strong>SocArXiv</strong> is a free, open-access preprint repository for the social sciences. It was founded in 2016 by sociologist Philip N. Cohen in partnership with the nonprofit <a href='/guides/center-for-open-science'>Center for Open Science (COS)</a> and runs on the <a href='/dictionary/term/open-science-framework-osf'>Open Science Framework (OSF)</a> — the same underlying infrastructure that hosts <a href='/dictionary/term/osf-preprints'>OSF Preprints</a> and sibling branded servers such as <a href='/dictionary/term/psyarxiv'>PsyArXiv</a> (psychology) and engrXiv (engineering). SocArXiv launched alongside those two services on December 5, 2016, as part of COS's branded-server model, in which a scholarly community operates its own named, separately moderated preprint service on shared OSF backend infrastructure rather than building hosting from scratch.</p> <p>SocArXiv is not a COS-run editorial operation in the way a journal is. It is governed by a volunteer steering committee of scholars and library professionals under the banner "SocOpen," and since 2021 its institutional home has been the University of Maryland Libraries, which provides the administrative and legal footing for the service as an ongoing academic unit rather than a standalone nonprofit.</p> <h2>Operational definition</h2> <p>A deposit is properly identified as a "SocArXiv preprint" if it meets all of the following:</p> <ul> <li>It was submitted through the SocArXiv service on OSF (osf.io/preprints/socarxiv), not simply cross-listed or mirrored from another repository such as <a href='/guides/arxiv-preprints-what-they-are-and-how-to-use-them'>arXiv</a> or SSRN.</li> <li>It falls within sociology or a closely related social-science field — SocArXiv accepts working papers, preprints, published-paper copies, data, and code across the social sciences broadly, not sociology alone.</li> <li>It has passed SocOpen's moderation screen, which is a scope-and-completeness check, not peer review: moderators confirm the item is scholarly, falls within a supported research area, is plausibly categorized, is correctly attributed to its stated authors, is in a moderated language, and is submitted in a text-searchable format (PDF or DOCX, not a scanned image).</li> <li>It has received an automatically assigned, persistent <a href='/dictionary/term/doi'>DOI</a> on upload, making it citable independent of eventual journal publication; a prior DOI issued by another publisher can also be linked to the same record.</li> <li>It carries one of SocArXiv's permitted reuse terms — CC BY 4.0, CC0 (public-domain dedication), or no license — selected by the author at deposit.</li> </ul> <h2>What SocArXiv is not</h2> <p>Posting to SocArXiv is explicitly not peer review, and a SocArXiv record is not, by itself, evidence that a manuscript has been formally accepted anywhere. Moderation checks scope and completeness, not scientific merit, methodology, or novelty. SocArXiv is also free to authors and readers — there is no submission fee or paywall, which distinguishes it from many venues that require an <a href='/dictionary/term/article-processing-charge'>article processing charge (APC)</a> at the journal stage.</p> <h2>Examples</h2> <ul> <li>A sociologist deposits a working paper analyzing survey data on labor-market outcomes to SocArXiv ahead of submitting it to a peer-reviewed journal, receiving a citable DOI immediately and using that DOI in grant reports and CVs while the manuscript is still under review elsewhere.</li> <li>A research team posts the accepted-but-not-yet-typeset version of a political-science article to SocArXiv under a publisher-permitted embargo, satisfying a funder's open-access mandate without waiting for the journal's own publication date.</li> </ul> <h2>Counter-example</h2> <p>A working paper in economics or sociology posted only to SSRN, a university's own institutional repository, or a personal website is <em>not</em> a "SocArXiv preprint," even though it may cover an identical topic and serve a similar function — the term refers specifically to material submitted through and moderated by the SocArXiv service on OSF, not to social-science preprints in general (for that broader category, see <a href='/dictionary/term/preprint'>preprint</a>).</p> <h2>Related terms</h2> <ul> <li><a href='/dictionary/term/preprint'>Preprint</a></li> <li><a href='/dictionary/term/osf-preprints'>OSF Preprints</a></li> <li><a href='/dictionary/term/psyarxiv'>PsyArXiv</a></li> <li><a href='/dictionary/term/open-science-framework-osf'>Open Science Framework (OSF)</a></li> <li><a href='/dictionary/term/doi'>DOI</a></li> <li><a href='/guides/how-to-choose-a-preprint-server'>How to Choose a Preprint Server</a></li> <li><a href='/guides/preprint-servers-explained'>Preprint Servers Explained</a></li> </ul>
TGA CTN and CTA (formerly CTX) Schemes
The CTN (Clinical Trial Notification) and CTA (Clinical Trial Approval, formerly called CTX -- Clinical Trial Exemption) schemes are the two regulatory pathways Australia's Therapeutic Goods Administration (TGA) uses to permit the lawful supply of an 'unapproved' therapeutic good -- one not yet entered on the Australian Register of Therapeutic Goods (ARTG) -- for use in a clinical trial. A trial is a CTN trial if the sponsor self-assesses the product and trial as suitable to proceed on notification alone: the sponsor lodges a CTN with the TGA (via the TGA Business Services online CTN form) after the trial protocol has HREC (Human Research Ethics Committee) approval, pays the CTN fee, and the trial may then start -- the TGA does not review the scientific data before the trial begins. A trial is a CTA trial if the sponsor instead applies for the TGA's formal evaluation and approval of the unapproved good before the trial can start; the TGA reviews the supporting data (quality, safety, and available efficacy information) and must approve the application before recruitment begins. Both schemes sit alongside, not instead of, HREC ethical review and (for CTN trials) the relevant institution's own governance authorisation -- TGA approval or notification concerns lawful supply of the unapproved product, not the ethics of the study design.
TCET (Transitional Coverage for Emerging Technologies)
TCET (Transitional Coverage for Emerging Technologies) is a CMS coverage pathway, finalized August 2024 (final procedural notice CMS-3421-FN, effective August 12, 2024), that expedites Medicare national coverage for a limited set of FDA-designated Breakthrough Devices by running the device through CMS's existing National Coverage Determination (NCD) and Coverage with Evidence Development (CED) machinery on an accelerated, pre-planned timeline rather than creating a new coverage authority. A device becomes TCET-eligible only if it holds FDA Breakthrough Device designation, has received FDA market authorization, falls within an existing Medicare benefit category, and is not already the subject of an NCD; CMS finalized an exclusion for digital-only technologies and capped the pathway at five new candidates accepted per year, reviewed on a quarterly cycle. A manufacturer engages CMS well before authorization: a non-binding Letter of Intent roughly 18-24 months out, formal self-nomination roughly 12 months out, a CMS-contractor-generated evidence preview identifying gaps in the existing clinical evidence, and a manufacturer-authored Evidence Development Plan (EDP) mapping how those gaps will be closed. If accepted, CMS pursues an NCD conditioned on CED, giving the device national Medicare coverage for several years -- potentially up to roughly five -- while the manufacturer collects the additional outcomes data the evidence preview identified as missing.
PMDA Clinical Trial Notification (CTN)
A Clinical Trial Notification (CTN, chiken todokede) is the regulatory filing a sponsor must submit to Japan's Ministry of Health, Labour and Welfare (MHLW) -- reviewed on MHLW's behalf by the Pharmaceuticals and Medical Devices Agency (PMDA) -- before initiating a clinical trial of an unapproved drug, or a new indication/dosage of an approved drug, in Japan. It is Japan's functional counterpart to a U.S. FDA Investigational New Drug (IND) application, but operates as a notification-and-review-window system rather than an affirmative approval: for an initial CTN, PMDA has a 30-day review period after receipt in which it can raise questions or object; if PMDA raises no objection within that window, the sponsor may proceed to start the trial. A submission counts as a CTN when it is filed under Japan's PMD Act (Act on Securing Quality, Efficacy and Safety of Products Including Pharmaceuticals and Medical Devices) and includes the trial protocol, investigational product information, and details of the investigators and institutions where the trial will be conducted.
Patent Pool
A patent pool is an arrangement among two or more independent patent holders -- which can include a university technology transfer office -- to combine a defined set of patents into a single licensing package administered through one licensor or pool entity, so a licensee can obtain rights to every pooled patent in one transaction, with royalties divided among members by a pre-agreed allocation formula. Distinct from a standard bilateral exclusive or non-exclusive license, which involves a single licensor.
What you can do here
Browse · search · download · cite · contribute
Browse
By domain or letter
All 1242 entries, grouped by their thematic domain or alphabetically. Filter by status and CRIS-readiness.
Learn moreSearch
Full-text + facets
Search definitions, picklists, sample values, and crosswalks. Type-ahead suggestions across all entries.
Learn moreDownload
Six open formats
JSON-LD · CSV · OWL · RDF Turtle · BibTeX · XLSX. Suitable for CRIS ingestion or local vocabulary servers.
Learn moreCite
Citable URIs + DOIs
Every entry has a stable URI. The dictionary itself has a Dataset DOI (assigned per release).
Learn moreRelated areas of CASRAI
Where to go next
The dictionary sits inside a broader stewardship programme. These adjacent surfaces extend the same vocabulary into roles, adoption tracking, crosswalks, governance, and source bibliography.
/credit
CRediT taxonomy
The 14 contributor roles (conceptualization, methodology, software, writing — original draft, etc.) and the canonical role definitions.
/credit/adoption
Publisher CRediT adoption tracker
Which journals and publishers ingest CRediT taxonomy contributor roles, with adoption status and submission-system support.
/federation/cross-walks
Crosswalks
Field-level mappings from CASRAI entries to DataCite, MARC, ORCID, Schema.org, and other vocabularies.
/about/working-groups
Working groups
The 20 stewardship working groups that draft, review, and ratify dictionary entries on a rolling release cadence.
/standards/releases
Release history
Per-release changelog of added, modified, and deprecated entries since the 2026 revival of CASRAI stewardship.
/resources/bibliography
Foundational bibliography
Peer-reviewed and grey-literature sources underpinning the dictionary entries, organised by domain and theme.
About
What the dictionary is
Provenance, governance, scope, and how the dictionary relates to NISO / euroCRIS / CODATA standards.
Contribute
Propose a term
Working groups review proposals during each release cycle. Community contributions get CRediT attribution.
Releases
v2026.1 release notes
Added, modified, and deprecated entries since the previous release. Versioned and citable.
For implementers and citers
Open, machine-readable, freely reusable
CASRAI Dictionary content is published under CC-BY 4.0. Use it in your CRIS, repository, publisher submission system, CRediT-aware editorial workflow, or in your own controlled-vocabulary infrastructure.
Frequently asked
Dictionary FAQ
- What is the CASRAI Dictionary?
The CASRAI Dictionary is a curated, citable, machine-readable controlled vocabulary for research-administration information. v2026.1 contains 1,242 entries spread across 24 thematic domains, each entry with a stable URI, structured picklists, and Schema.org
DefinedTermmarkup. See about the dictionary for scope and provenance.- How is the Dictionary structured?
The Dictionary is partitioned into 24 thematic domains grouped under 5 stewardship tracks: contributorship, identifiers and systems, computational, governance and integrity, and lifecycle and equity. Each domain is stewarded by a community working group that drafts, reviews, and ratifies its entries. Browse the domain map at /dictionary, or jump to /dictionary/browse for the A-Z view.
- Can I propose a new term?
Yes. Working groups review proposals during each release cycle, and community contributions are credited via CRediT-style attribution in the changelog. Open a proposal at /dictionary/contribute; the relevant working group will route it through the review process for the next versioned release.
- Under what licence are the definitions released?
The CASRAI Dictionary is published under the Creative Commons Attribution 4.0 International licence (CC-BY 4.0). You may reuse, redistribute, and adapt the content commercially or non-commercially, provided you give appropriate credit. Implementation patterns are documented under /implement.
- How do I cite the Dictionary?
Each Dictionary release is citable as a Dataset with a versioned URI (e.g.
https://casrai.org/dictionary/for v2026.1) and per-entry stable URIs athttps://casrai.org/dictionary/term/<slug>. The cite page provides citation strings in APA, BibTeX, and RIS formats.- Where do the legacy CASRAI terms come from?
CASRAI was founded in 2006 and built an extensive legacy dictionary before the 2020 stewardship handover. The 2026 revival re-imported and re-published the legacy entries under their original domains, retaining provenance metadata. The current 1,242-entry corpus combines legacy holdings with newly drafted entries. See /about/history for the timeline and /dictionary/changelog for per-release migration notes.
- How many terms are in the CASRAI Dictionary?
The CASRAI Dictionary currently contains 1,242 terms in release v2026.1, distributed across 24 thematic domains grouped under 5 stewardship tracks. Each term has a stable URI, structured definition, optional picklists, and Schema.org
DefinedTermmarkup. The next release, v2026.2 in November 2026, expands the corpus with additional sustainability and AI-disclosure entries. See /dictionary/changelog.- Is the CASRAI Dictionary free?
Yes, the CASRAI Dictionary is free to use. It is published under the Creative Commons Attribution 4.0 International licence (CC-BY 4.0), with no paywall, no registration, and no API spend for read access. You may reuse, redistribute, translate, and bundle it commercially or non-commercially, provided you give appropriate credit to CASRAI.
- How do I cite a dictionary term?
To cite a dictionary term, use its stable URI at
https://casrai.org/dictionary/term/<slug>together with the dictionary release version. Each term page emits a citation widget producing APA, BibTeX, RIS, and Chicago forms on the fly. To cite the release as a whole, use the Zenodo DOI listed on /dictionary/cite.
