CRediT Taxonomy Form: From Submission Checkbox to Crossref Metadata

The credit taxonomy form is the checkbox interface built into manuscript submission systems — such as Editorial Manager and ScholarOne — that lets a corresponding author assign one or more of the 14 CRediT contributor roles to each author. That selection is not just cosmetic: it is designed to travel, unchanged, from the submission form into the article’s JATS XML and, increasingly, into the DOI metadata that Crossref exposes to the wider scholarly record.

CRediT (Contributor Roles Taxonomy) is a standardised, machine-readable vocabulary of 14 contribution types — such as Conceptualization, Data curation or Writing – original draft — used to describe what each author or contributor actually did, distinct from the separate question of who qualifies for authorship. CASRAI originated the CRediT contributor role taxonomy in 2014. The standard is now stewarded by NISO as ANSI/NISO Z39.104-2022, with descriptors available in multiple languages.

This article is written for developers, editorial-systems teams and publishing operations staff who need to understand the full pipeline — not just what CRediT means, but how a checkbox selection becomes a permanent, queryable metadata field.

• What is the CRediT taxonomy form?
• How manuscript systems capture the form at submission
• How CRediT roles are encoded in JATS XML
• How CRediT metadata reaches Crossref and the DOI record
• Frequently asked questions
• Implications for publishers and developers
• What happens next

What is the CRediT taxonomy form?

The CRediT taxonomy form is the structured data-entry step, embedded in a journal’s manuscript submission system, where the corresponding author declares each contributor’s role from a fixed list of 14 options. It is a controlled-vocabulary form, not free text — every selection maps to one of the standard CRediT terms, each with its own stable identifier at credit.niso.org.

The 14 roles captured by the form are:

• Conceptualization
• Data curation
• Formal analysis
• Funding acquisition
• Investigation
• Methodology
• Project administration
• Resources
• Software
• Supervision
• Validation
• Visualization
• Writing – original draft
• Writing – review & editing

A contributor can hold more than one role, and a single role can be assigned to more than one contributor. Some submission systems also let authors record a degree of contribution — lead, equal or supporting — but this qualifier is optional and is not currently part of the CRediT standard itself; individual publishers decide whether to collect it.

How manuscript systems capture the form at submission

CRediT capture happens earliest in the pipeline, at manuscript submission, before peer review begins. Aries Systems’ Editorial Manager has supported CRediT since 2016 and Clarivate’s ScholarOne Manuscripts since 2018 — the two dominant commercial manuscript submission systems (MSSs) used across most major publishers. Open Journal Systems (OJS), the leading open-source platform, offers CRediT as a plugin.

During submission, the corresponding author works through the contributor list and ticks the applicable CRediT roles for each name. The system typically requires this step to be completed, or at least reviewed, before the manuscript can proceed to the editor. Best practice — reflected in NISO’s implementation guidance — is for all named authors to review and confirm the roles assigned to them before the form is finalised, reducing later disputes over credit.

How CRediT roles are encoded in JATS XML

Once accepted, the article moves into production, where the checkbox selections are converted into JATS XML (Journal Article Tag Suite) — the NISO standard (ANSI/NISO Z39.96) used to structure scholarly article content and metadata for archiving, indexing and reuse. This is the step that turns a form selection into genuinely machine-readable data rather than just display text on the published page.

The JATS4R (JATS for Reuse) working group, in a recommendation published 13 September 2021 and current at version 1.2, specifies exactly how CRediT terms must be tagged inside the <role> element within <contrib-group>/<contrib>. Under JATS 1.2 and above, each role carries four attributes:

• @vocab — must always carry the value “credit”
• @vocab-identifier — must point to https://credit.niso.org/
• @vocab-term — must identify the specific CRediT term (e.g. “Conceptualization”)
• @vocab-term-identifier — must point to the URL of that specific term (e.g. https://credit.niso.org/contributor-roles/conceptualization/)

Each CRediT term requires its own <role> element — where a contributor holds multiple roles, the element is repeated rather than combined. JATS4R also provides a validator tool that flags mismatches, such as a role term with no corresponding @vocab="credit" attribute. Publishers still using the older JATS 1.1 DTD can tag CRediT terms using the @content-type attribute instead, though JATS4R recommends migrating to 1.2 for more robust support.

How CRediT metadata reaches Crossref and the DOI record

The final leg of the pipeline is deposit with Crossref, the DOI registration agency that most scholarly publishers use to register article identifiers and metadata. Per Crossref’s published strategy roadmap, Crossref is working on metadata ingest schema v.5.5 to add native support for the CRediT taxonomy of contributor roles — meaning the same structured role data captured in JATS XML can be deposited directly against the article’s DOI record.

Once that ingest path is live, CRediT information becomes queryable through Crossref’s public APIs rather than sitting only inside a publisher’s own XML or HTML. That has knock-on value for research information systems, ORCID contribution records and institutional CRIS platforms, all of which can pull structured contribution data by DOI instead of scraping an “Author Contributions” paragraph from the rendered article page.

Note that CRediT is explicitly not designed to determine authorship. ICMJE’s authorship criteria and COPE’s guidance on authorship disputes remain the governing frameworks for who qualifies as an author; CRediT — and the form that captures it — only describes what each named contributor did.

Frequently asked questions

What is the CRediT taxonomy?

CRediT is a standardised system of 14 roles for describing individual contributions to a research output. When an author submits to a publisher using CRediT, the submission form asks them to select the applicable roles for each contributor, rather than relying on free-text acknowledgements.

What are the 14 roles of the CRediT taxonomy?

The 14 roles are Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing – original draft, and Writing – review & editing — each with a stable identifier at credit.niso.org.

How do I CRediT someone in a research paper?

Agree contributor roles with co-authors before submission, then use the journal’s CRediT taxonomy form to assign one or more of the 14 roles to each name. Where a journal has no integrated form, contributions can be listed manually in an author contribution statement or acknowledgement section.

Implications for publishers and developers

For manuscript-system developers, the practical requirement is straightforward but easy to get wrong: the checkbox output must map one-to-one onto the JATS4R attribute set, not a locally invented tag scheme, or the resulting XML will fail JATS4R validation and lose machine-readability further down the chain. For publishing operations teams, the Crossref v.5.5 schema update is the point at which CRediT stops being a publisher-side display feature and becomes part of the permanent scholarly record attached to the DOI itself.

Institutions and funders tracking research contribution data — for hiring, promotion or grant-reporting purposes — should treat Crossref-deposited CRediT metadata, once available, as a more reliable source than scraping “Author Contributions” text from HTML pages, since the underlying <role> tagging is what the form was designed to produce in the first place.

What happens next

As Crossref’s schema v.5.5 rolls out, expect wider adoption of CRediT deposit among publishers that have not yet enabled it, and closer scrutiny from JATS4R’s validator tooling as more XML pipelines attempt to encode CRediT roles at scale. The direction of travel is consistent: from a checkbox on a submission form, through a standard XML attribute set, to a queryable field on a permanent DOI record. Developers building or auditing manuscript-system integrations should treat the CRediT taxonomy and its individual contributor roles as the canonical reference points, alongside the broader question of how contribution data relates to authorship criteria.