CRediT in JSON-LD

A minimal author + roles object

Schema.org’s conformant way to attach a CRediT role to an author is the Role wrapper. roleName is a property of Role (its range is Text — the human-readable role name), not a property of Person. So the author is defined once as a Person whose @id is their ORCID iD, and each contribution is a Role object in a contributor array that references that author back by @id. The url on each Role resolves it to one of the 14 canonical role pages, and because every Role points at the same @id, a crawler resolves a single author entity rather than duplicating the person per role.

CreativeWork with a contributor[] of Role objects

json

{
  "@context": "https://schema.org",
  "@type": "ScholarlyArticle",
  "headline": "A note on reproducibility in clinical audit",
  "author": {
    "@type": "Person",
    "@id": "https://orcid.org/0000-0002-1825-0097",
    "name": "Dr Jordan Avery",
    "url": "https://casrai.org/credit/authors/jordan-avery",
    "sameAs": [
      "https://orcid.org/0000-0002-1825-0097",
      "https://casrai.org/credit/authors/jordan-avery"
    ]
  },
  "contributor": [
    {
      "@type": "Role",
      "roleName": "Writing – original draft",
      "url": "https://casrai.org/credit/roles/writing-original-draft",
      "contributor": { "@id": "https://orcid.org/0000-0002-1825-0097" }
    },
    {
      "@type": "Role",
      "roleName": "Conceptualization",
      "url": "https://casrai.org/credit/roles/conceptualization",
      "contributor": { "@id": "https://orcid.org/0000-0002-1825-0097" }
    }
  ]
}

The CASRAI Verified Contributor Directory publishes each member as a Person with their roles expressed as Role objects using exactly this encoding, so an ORCID-verified profile is machine-readable to the same crawlers and discovery platforms that consume the markup above. The same pattern also underpins the CASRAI CRediT Connector plugin, which emits this exact markup on self-hosted WordPress journals and blogs.

Per-role DefinedTerm

Each CRediT role page on this site emits a stand-alone DefinedTerm JSON-LD payload. This is what Google's Rich Results Test consumes for indexing.

Standalone DefinedTerm for the Investigation role

json

{
  "@context": "https://schema.org",
  "@type": "DefinedTerm",
  "@id": "https://casrai.org/credit/roles/investigation",
  "name": "Investigation",
  "identifier": "https://casrai.org/credit/roles/investigation",
  "description": "Conducting a research and investigation process, specifically performing the experiments, or data/evidence collection.",
  "inDefinedTermSet": "https://casrai.org/credit#taxonomy",
  "url": "https://casrai.org/credit/roles/investigation",
  "license": "https://creativecommons.org/licenses/by/4.0/"
}

DefinedTermSet for the full CRediT vocabulary

The 14-role taxonomy as a DefinedTermSet

json

{
  "@context": "https://schema.org",
  "@type": "DefinedTermSet",
  "@id": "https://casrai.org/credit#taxonomy",
  "name": "Contributor Roles Taxonomy (CRediT)",
  "description": "14 standardised contributor roles for scholarly research outputs. ANSI/NISO Z39.104-2022.",
  "url": "https://casrai.org/credit",
  "license": "https://creativecommons.org/licenses/by/4.0/",
  "version": "ANSI/NISO Z39.104-2022",
  "identifier": "Z39.104-2022",
  "hasDefinedTerm": [
    { "@id": "https://casrai.org/credit/roles/conceptualization" },
    { "@id": "https://casrai.org/credit/roles/data-curation" },
    // ... 12 more
  ]
}

For article-level XML pipelines, the same 14 roles are encoded as JATS <role> elements rather than JSON-LD.