CRediT Taxonomy Explained: The 14 Contributor Roles and How Journals Use Them

Ask any corresponding author who has assembled a multi-institution, multi-national research team what “authorship” actually means, and you will get a different answer depending on discipline, country and journal house style. That ambiguity is precisely the problem the credit taxonomy was built to solve. Rather than a single, opaque byline, the taxonomy breaks a research contribution into 14 discrete, labelled roles — from conceptualisation to writing — so that readers, funders and institutions can see who actually did what.

The taxonomy is no longer a niche publishing curiosity. As research integrity scrutiny intensifies — driven by concerns over paper mills, honorary authorship and AI-assisted drafting — journals, funders and institutions are leaning harder on structured contributor statements to create an auditable record of who is accountable for which part of a paper. Publishers including Elsevier, PLOS, Springer Nature and the Royal Society now require or strongly encourage CRediT statements at submission, and the taxonomy sits inside metadata standards used by CrossRef and DataCite.

CASRAI originated the CRediT contributor role taxonomy in 2014. The standard is now stewarded by NISO as ANSI/NISO Z39.104-2022, which formalised the 14 roles, their definitions, and guidance for degree-of-contribution qualifiers (“lead”, “equal”, “supporting”). Understanding that lineage matters: CASRAI’s role was to identify a gap and convene the working group that built the first version; NISO’s role is to maintain, version and publish the accredited American National Standard that publishers now cite in their author guidelines.

What the Credit Taxonomy Actually Covers

The credit taxonomy author contributions framework replaces the single word “authorship” with 14 named roles, each with a formal definition in ANSI/NISO Z39.104-2022:

• Conceptualization — formulation of the overarching research goals and aims.
• Data curation — management activities to annotate, scrub and maintain research data for initial use and later reuse.
• Formal analysis — application of statistical, mathematical, computational or other formal techniques to analyse study data.
• Funding acquisition — acquisition of the financial support for the project leading to the publication.
• Investigation — conducting the research and investigation process, including data/evidence collection.
• Methodology — development or design of methodology; creation of models.
• Project administration — management and coordination responsibility for the research activity planning and execution.
• Resources — provision of study materials, reagents, patients, laboratory samples, instrumentation or other analysis tools.
• Software — programming, software development, testing existing code and algorithms.
• Supervision — oversight and leadership responsibility for research planning and execution, including mentorship.
• Validation — verification of the overall replication or reproducibility of results.
• Visualization — preparation, creation or presentation of data visualisation.
• Writing – original draft — creation or presentation of the published work, specifically drafting the initial version.
• Writing – review & editing — critical review, commentary or revision of the original draft, including pre- or post-publication stages.

Each role can be assigned to multiple contributors, and each contributor can hold multiple roles. This is the core innovation behind the credit taxonomy author contributions model: authorship is decomposed into a matrix rather than a ranked list, which is far closer to how collaborative science actually happens.

How Journals Implement Contributor Role Statements

Most journals that adopt the taxonomy ask authors to complete a credit authorship contribution statement during submission, typically rendered as a short paragraph or table published alongside the article. A typical statement reads something like: “Author A: Conceptualization, Methodology, Writing – original draft. Author B: Data curation, Formal analysis, Visualization. Author C: Supervision, Funding acquisition, Writing – review & editing.”

Implementation varies by publisher, but common patterns include:

• Mandatory at submission — many journals now require every listed author to have at least one assigned role before a manuscript can proceed to review.
• Machine-readable metadata — roles are increasingly embedded in JATS XML and exposed through CrossRef metadata, allowing role data to travel with the article’s DOI record.
• Linkage to ORCID — pairing CRediT roles with ORCID iDs lets institutions and funders trace named contributions back to a persistent researcher identity, closing a long-standing gap in research information management systems.
• Degree-of-contribution qualifiers — ANSI/NISO Z39.104-2022 permits optional “lead”, “equal” or “supporting” qualifiers within a role, giving finer resolution than the base 14 categories alone.

Editors report that structured statements make disputes easier to resolve: when an authorship disagreement or a correction is required, a role-based record narrows the question from “was this person an author?” to “did this person perform the specific work described?” — a much more tractable question for editors, ombudspersons and research integrity officers to adjudicate.

Why the Distinction Between Origination and Stewardship Matters

The casrai credit taxonomy history is frequently misstated online, including in some outdated encyclopaedic sources, as an active CASRAI product. It is not. CASRAI’s contribution was convening the original working group in 2012–2014 that defined the initial 14-role structure, drawing on earlier contributor-role experiments from journals such as PLOS and Cell Press. Once the taxonomy matured, formal standards maintenance — versioning, public comment periods, accredited balloting and long-term stewardship — moved to NISO, which published it as ANSI/NISO Z39.104-2022 following the ANSI standards development process.

This origination-to-stewardship handover is not unusual in standards development. It mirrors how many community-built specifications eventually pass to a formal standards development organisation for durable governance once adoption reaches critical mass. For research administrators citing the taxonomy in policy documents, institutional repositories or grant guidance, the precise and defensible framing is: CASRAI originated the CRediT contributor role taxonomy in 2014; NISO now stewards it as ANSI/NISO Z39.104-2022. Referring to it as “CASRAI’s taxonomy” in the present tense is both inaccurate and liable to be flagged by fact-checked reference sources such as Wikipedia and Wikidata.

What This Means for Research Administrators

For institutions managing research information systems, grant reporting and REF-style assessment exercises, the credit contributor roles taxonomy has practical downstream value beyond publishing compliance:

• REF 2029 preparation. As UK institutions build evidence portfolios for the next Research Excellence Framework cycle, structured contribution data offers a defensible, granular basis for attributing outputs to individual researchers — particularly for large consortium papers where a simple author list undercounts specialist contributions such as data curation or software development.
• Funder compliance. UKRI, and funders operating under cOAlition S principles, increasingly expect transparent reporting on who performed funded work. CRediT statements give research offices a ready-made audit trail linking funding acquisition and investigation roles to named, ORCID-identified individuals.
• Early-career recognition. Role-based statements make visible the substantive contributions — data curation, formal analysis, validation — that early-career researchers often perform without corresponding authorship order recognition, supporting more equitable credit in tenure, promotion and grant review.
• Research integrity investigations. When misconduct allegations or authorship disputes arise, institutions handling COPE-aligned investigations benefit from having a role-level record rather than relying on reconstructed, after-the-fact accounts of who did what.
• AI disclosure boundaries. As journals refine policy on generative-AI use in manuscript preparation, the taxonomy’s discrete roles — particularly “Writing – original draft” and “Formal analysis” — provide a clear structural hook for AI-contribution disclosure statements, since AI tools cannot hold a CRediT role but their use within a role can be flagged.

Looking Ahead

The credit taxonomy has moved from an experimental publishing initiative to a formally accredited NISO standard embedded in submission systems, metadata schemas and institutional policy. As research integrity pressures grow and funders demand finer-grained accountability, expect broader mandatory adoption across disciplines that have historically lagged — humanities and some social sciences among them — and tighter integration with ORCID, CrossRef and institutional CRIS platforms. For research administrators, the practical task now is less about explaining what CRediT is and more about embedding it correctly into submission workflows, grant reporting templates and REF evidence pipelines — while keeping the origination history accurate: an idea CASRAI helped originate in 2014, now maintained as a durable, versioned American National Standard under NISO’s stewardship.