Contributor Roles Taxonomy

The Contributor Roles Taxonomy, commonly known as CRediT, is a controlled vocabulary of types of contributions to a research project.^[1] CRediT is commonly used by scientific journals to provide an indication of what each contributor to a project did. The CRediT standard includes machine-readable metadata.^[2]

Historically, articles in scientific journals included a list of authors but gave little or no indication of what each author did. Beginning in the 1990s, Drummond Rennie and others argued that journals should indicate who did what.^[3] In 2012, a first draft of the CRediT standard was developed in the context of biomedicine by a group of funders, publishers, and researchers.^[1] In 2022, CRediT became an ANSI/NISO standard^[4] (ANSI/NISO z39104-2022-credit).^[5] The official CRediT standard is in English only, but translations into other languages have been created.^[6]

The CRediT standard provides 14 categories of contribution:

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

NISO also provides a longer description of each category.^[7]

Of the publishers that use CRediT, some have also implemented the possibility for researchers to indicate whether a researcher played a lead, equal or supporting role for each of their contribution types.^[8]

Researchers submitting an article for publication to a journal using CRediT are typically asked to tick a box next to each of the fourteen contribution types associated with the work that they did.

Possible benefits of indicating who did what using CRediT are:

• Better assignment of credit for scientific work.^{[1]: 151 [9]}
• Fewer disputes among potential authors.^{[1]: 151 [9]}
• An increase in the incentives for collaboration.^{[1]: 151 [9]}
• An increase in the incentives for sharing of data and code.^[1]: 151
• Increasing accountability when questions arise about particular aspects of a project, through its indication of which researcher(s) did the associated work^[3]
• Enabling large-scale analyses of the scientific literature to improve understanding of how science works^[10][11]

Citing inadequacies with current practices in listing authors of papers in medical research journals, Drummond Rennie and co-authors, writing in the Journal of the American Medical Association (JAMA) in 1997, called for:

a radical conceptual and systematic change, to reflect the realities of multiple authorship and to buttress accountability. We propose dropping the outmoded notion of author in favor of the more useful and realistic one of contributor.^[1]: 152

In the 2000s, prestigious journals such as Nature began requiring authors to provide information about what their contributions were,^[12] but there was no widely-used or machine-readable standard for this.

In 2012, a draft taxonomy was created at a workshop held at Harvard involving biomedical scientists, publishers, and research funders.^[13][14] Beginning in 2014, a working group of publishers, funders, and university representatives began meeting to refine the draft of the CRediT taxonomy, coordinated by the Consortia for Advancing Standards in Research Administration Information (CASRAI).^[13]

By 2017, PLOS journals and eLife had adopted CRediT,^[15][16] and in 2018 it was endorsed by representatives of the National Academy of Sciences.^[17] Over the next several years, some of the largest publishers of scientific journals began using CRediT.^{[18][19][20][21]}

CRediT was primarily designed with scientific journal articles in mind,^[22] and even within that, some researchers have reported difficulty mapping the CRediT categories onto their field.^[23] In 2023, a systematic scoping review identified 20 unique ethical issues related to contributor role taxonomies like CRediT.^[24] One of the highlighted issues pertains to the use of taxonomies like CRediT in specific contexts. For example, the work of performing a literature review may arguably be classified as "Investigation" or "Formal analysis", but neither may seem a good fit.^[25] In a study of one psychology research project, independent researchers read detailed descriptions of other researchers' contributions, the results indicated that the independent researchers had low agreement about both the number and type that the contributions should be classified into.^[26] Nevertheless, there have also been suggestions on how to extend CRediT roles in three phases including "Identification of candidate roles", "Deciding what roles to include in the standard lists", and "Integrating new roles into the existing list of roles".^[27]

CrossRef, which is the largest maintainer of metadata about scientific articles, does not currently support import of CRediT information, but this feature is in development.^[28]

As the International Committee of Medical Journal Editors has pointed out, documenting contributions with CRediT or another scheme "leaves unresolved the question of the quantity and quality of contribution that qualify an individual for authorship", suggesting that authorship guidelines are still necessary, although authorship guidelines also typically fail to specify the quantity of contribution reqiured.^[29]

The Contributor Roles Ontology is an extension of the CRediT taxonomy into more specific roles.^[30] An extension for clinical trials (CRediT-RCT) has been proposed.^[31]

Other taxonomies have been created that may be more suitable to other fields, such as the Taxonomy of Digital Research Activities in the Humanities (TaDiRAH).^[32]