CRediT & Authorship Attribution

CRediT (the Contributor Roles Taxonomy) defines 14 standardized roles — such as Conceptualization, Data Curation, and Writing – Original Draft — that describe what each person contributed to a research output. It is distinct from authorship: CRediT roles document contribution, while separate criteria such as ICMJE’s four authorship conditions determine who qualifies to be listed as an author.

CASRAI originated CRediT during 2012–2014 development work, and now stewards it jointly with NISO through the NISO CRediT Standing Committee. The taxonomy is formalized as ANSI/NISO Z39.104-2022. For the full definition and standardization history, start with the CRediT overview page.

What CRediT actually standardizes

CRediT is a controlled vocabulary of 14 contributor roles. Each author on a research output can be assigned one or more roles, and many publishers now support an optional degree qualifier (lead, equal, or supporting) per role. The 14 roles are:

• Conceptualization
• Data Curation
• Formal Analysis
• Funding Acquisition
• Investigation
• Methodology
• Project Administration
• Resources
• Software
• Supervision
• Validation
• Visualization
• Writing – Original Draft
• Writing – Review & Editing

Each role page covers the formal definition, when the role applies, how to encode it (JATS XML, JSON-LD, Crossref deposit), and discipline-specific examples — that depth lives on the individual role pages, not duplicated here.

CRediT is not an authorship test

A recurring point of confusion: CRediT describes contribution, it does not decide who counts as an author. That determination is separately governed by criteria such as the ICMJE authorship criteria — a four-part test covering substantial contribution, drafting or critically revising the work, final approval of the version to be published, and accountability for accuracy and integrity. A person can appear in a CRediT statement for a limited role (for example, Resources or Funding Acquisition) without meeting a journal’s authorship bar, and conversely, meeting the authorship bar doesn’t specify which of the 14 roles that person actually performed. Using CRediT alongside — not instead of — a journal’s authorship policy is the correct pattern. CASRAI has covered how this plays out in medical-research publishing specifically in CRediT statements in medical research: the current landscape.

Writing and submitting a CRediT statement

Most major publishers now collect CRediT roles at submission — either through structured fields in the manuscript system or a narrative paragraph rendered above the acknowledgments in the published version. For the mechanics of assigning roles and degree qualifiers, see:

• CRediT guidance for authors — step-by-step role assignment at submission
• Writing a CRediT statement — the author-facing walkthrough
• CRediT author statement samples — adaptable sample statements for common authorship scenarios, plus the important distinction that author order is not the same thing as a CRediT statement

For where CRediT is actually in use across scholarly publishing, see CRediT adoption, which tracks implementation level (native, available, or variable) by publisher.

Authorship problems CRediT doesn’t solve

Because CRediT documents contribution rather than adjudicating authorship, the taxonomy on its own doesn’t prevent — and isn’t designed to prevent — the authorship-integrity problems research offices actually deal with. CASRAI’s Dictionary covers the specific vocabulary for these:

• Gift authorship — crediting someone who did not meet authorship criteria
• Ghost authorship — omitting someone who did meet authorship criteria (or did the writing) from the author list
• Authorship dispute — disagreement over inclusion, order, or role attribution among contributors

These sit at the boundary of attribution and research integrity more broadly — CASRAI’s integrity and compliance content covers the misconduct-policy side of that boundary in more depth.

Contributors who aren’t authors

Not everyone who contributes to a research output qualifies as, or wants to be listed as, an author. CRediT and standard acknowledgment practice both make room for this distinction:

• Non-author contributor
• Acknowledged contributor
• Co-corresponding author convention — where more than one author shares corresponding-author responsibilities

CRediT, AI-assisted contributions, and authorship

As AI-assisted writing and analysis tools have become common in manuscript preparation, publishers and standards bodies (ICMJE among them) have been explicit that an AI tool cannot be listed as an author — it cannot take accountability for the work, which is a requirement of authorship criteria, not just a formality. CASRAI’s Dictionary tracks this vocabulary directly:

• AI as author
• AI co-authorship rejection (ICMJE 2023)
• Author responsibility (for AI use)
• Acknowledgement vs. authorship for AI

Connecting contributor identity: ORCID and CRediT

CRediT statements are most useful to downstream systems (CRIS platforms, funder reporting, institutional repositories) when each contributor is also identified unambiguously — which is where ORCID iD comes in. Machine-readable CRediT plus ORCID together are what let a role assertion travel with a specific, disambiguated person across systems, rather than living only as prose in a published paper.

Frequently asked

Does a CRediT statement replace an authorship policy?

No. A journal’s authorship policy (commonly built on the ICMJE criteria) still determines who is listed as an author at all; CRediT only records what each listed author (and in some publisher implementations, some non-author contributors) actually did.

Who decides which CRediT roles each author gets?

CASRAI does not adjudicate this — it is a matter for the author team to agree on, typically coordinated by the corresponding author, before or at submission. Disagreement over role assignment is a form of authorship dispute and is generally best resolved through the same institutional or editorial channels used for other authorship disagreements.

Is CRediT required by all journals?

No — adoption is broad but not universal, and implementation varies by publisher (native structured fields vs. a submitted narrative statement). See CRediT adoption for the current landscape by publisher.

Can more than one author hold the same CRediT role?

Yes. Multiple authors can be assigned the same role, and many publisher implementations let a shared role carry a degree qualifier (lead, equal, or supporting) to indicate relative contribution.

What’s on this hub, and what isn’t yet

This page indexes CASRAI’s existing CRediT and authorship-attribution content — the taxonomy overview, all 14 role pages, the author-statement guide, and the relevant Dictionary vocabulary. Guides and comparisons published under this cluster going forward (for example, on resolving authorship-order disputes, or CRediT for non-traditional contributors) will appear automatically below as they’re added.