Competing Interests: L.A. and V.K. were involved in the initial inception and development of the CRediT taxonomy. They now co-chair the NISO Standing Committee on CRediT

10 Years Ago, a Community of Researchers, Publishers, and Institutions Came Together to Try to Solve a Problem That Had Plagued Scholarly Publishing for Decades:

The Opacity of Authorship.

That effort was documented in the Report on the International Workshop on Contributorship and Scholarly Attribution, published in September 2012. Who actually did what in a published research study? Traditional author lists offered few clues, and the consequences (from honorary authorship to hidden contributions) were undermining both research integrity and fair recognition of researchers' work.

One practical solution was the creation of the Contributor Role Taxonomy, better known as CRediT, a straightforward system describing 14 distinct tasks that researchers typically perform (table). The taxonomy was built from a combination of commonly described roles in acknowledgments sections, free-text author contribution statements, and existing shorter taxonomies being used mainly by science, technology, engineering, and medical (STEM) publishers at the time.

See more examples for each CRediT role

A decade on, CRediT has achieved remarkable adoption, appearing in nearly 850,000 publications in 2024 alone, representing about 22% of all openly available scholarly output published in that year (figure). For those working in research integrity, this growth represents more than just better record-keeping; it demonstrates increasing commitment toward transparency in research that can support accountability and help to combat some of the most pressing threats to the scholarly record.

Number of publications including CRediT roles

Articles, preprints and conference papers where full text was available

What Problem Is CRediT Designed to Solve?

The origins of CRediT lie in frustrations with traditional authorship conventions that vary by discipline and often come from a time when research was done by small teams of researchers whose role could easily be inferred. As teams have grown, so have author lists, which now provide limited information about individual contributions. Author position alone is well known to be a poor proxy for contribution or responsibility. In fields like economics, alphabetical ordering is standard, while in STEM, as a result of the need for ‘team science’ and large collaborations, the phenomenon of "hyper authorship" can result in author lists of hundreds or even thousands of names.

This opacity creates fertile ground for questionable authorship practices, including "ghost authorship", where contributors are excluded from the author list, and "honorary authorship", where non-contributors are included. It is difficult to determine the extent of these practices, but some estimates suggest they occur in as many as 1/5 of biomedical papers. Questionable authorship practices are not only unfair, but they also undermine accountability. When conflicts of interest exist, unclear attribution can obscure responsibility. If results are later challenged, this lack of transparency can hinder investigations and may lead to accountability being shaped by power dynamics rather than actual contributions.

Greater visibility of individual contributions also supports more responsible research assessment. By documenting who did what, CRediT provides a clearer route for recognizing the full range of contributions that underpin modern, team-based research, beyond being first or last author (see table above). This is particularly important for early-career researchers, technical specialists (e.g., statisticians, software developers), and contributors working within large collaborations, where author position is a poor indicator of contribution or responsibility. In this way, contribution transparency aligns with wider efforts, such as DORA and CoARA, to move beyond simplistic authorship-based metrics toward fairer and more meaningful recognition of research contributions.

Research Integrity in the Spotlight

The accountability and transparency use cases for CRediT have intensified over the past decade as pressures on the research publishing system have grown. Publication volumes have increased sharply, driven not only by growing research activity but also by unhelpful career progression incentives, institutional performance metrics and business models. This environment reinforces a "publish or perish" culture, in which quantity is put over quality and peer review systems are further strained. Paper mills, the illicit operations that sell authorship on often fabricated articles, have proliferated in response to these perverse incentives and exploit weaknesses in peer review, credential checks and publisher workflows. The large-scale fraud enabled by these practices has led to increased scrutiny by publishers and sharply rising retraction rates, with over 10,000 papers retracted in 2023 alone. And as publishers' detection methods develop, there is increasing evidence of the potential scale of paper mill activity across the scientific literature.

Open science mandates have improved transparency, and open science practices can be a signal of a researcher’s confidence in their work and willingness to have it scrutinized. Preliminary analysis suggests that papers flagged for potential paper mill characteristics are less likely to share the underpinning data and code. This is not a reliable signal, nor is it likely to be long-lived, as paper mills have also shown the ability to produce highly convincing outputs, including fabricated data, manipulated images, and fictitious author lists.

However, mechanisms that make individual contributions and responsibility visible and explicit, if coupled with recognition and reward for the willingness to be held accountable, can support good scientific conduct. A change of incentives, including the reward of rigor and accountability over volume, will be critical to promote research integrity. Transparency of contributions to enable more responsible research assessment is an essential step in that direction.

CRediT cannot, on its own, resolve entrenched problems in research culture, but it can provide a practical foundation for strengthening good research practice and integrity. By making author contributions explicit, including responsibility for data, methods, analysis, and software, it increases transparency, accountability, and clarity around who did what, and supports research integrity in several concrete ways:

Evidence of Impact and Adoption

A retrospective analysis of CRediT's first decade shows rapid and sustained uptake, suggesting growing recognition of its value across the research community. Our analysis of articles in the Dimensions database (where full text is available) indicates that CRediT terms appeared in approximately 3.2 million publications between 2015 and 2024, with particularly strong growth in recent years, which was driven largely by the integration of CRediT into the major manuscript submission systems (e.g., Editorial Manager, Scholar One, and OJS), thus making it easier for publishers to capture CRediT roles within their journal workflows (see figure above).

CRediT has also been validated as a tool to support research integrity. The U.S. National Academies of Sciences has highlighted its role in improving transparency of author contributions in its 2017 report on fostering integrity in research, while UK initiatives on research culture and team science have encouraged its use. In 2022, CRediT was approved as a formal standard by the National Information Standards Organization (NISO) and the American National Standards Institute (ANSI). And in 2024, CRediT received a "Highly Commended" award in the UK Hidden REF Competition for its role in raising the profile of diverse research contributions.

Finally, CRediT data is already enabling large-scale meta-research on the research system itself, including being used to examine gender differences across research roles and contributions, demonstrating its value in building evidence to inform improvements in research culture and integrity.

Challenges That Remain

Like many community-led initiatives, CRediT has faced constraints due to limited dedicated resources for ongoing governance and management. As a result, while it is currently hosted by NISO, CRediT continues to rely on volunteer effort to raise awareness, support implementation, and steward the taxonomy, balancing stability with the need to keep roles aligned with evolving research practices.

Although integration of CRediT into publisher submission systems, alongside other author-based metadata such as ORCID iDs, has improved capture of contribution data, implementation remains uneven. Some publishers mandate CRediT, while others treat it as optional. In addition, the absence of CRediT from Crossref’s article metadata schema to date has led to variable data quality and limited its routine availability in bibliographic databases.

This fragmentation makes it harder for universities, funders, and policymakers to access CRediT data reliably and also complicates integrity checks, such as linking outputs, identifying patterns, and conducting reviews. While Crossref has indicated its intention to add CRediT in the near term, stronger governance and standardization will be important to fully realize its value. Even so, CRediT already provides meaningful support for research integrity, with benefits likely to increase as adoption and infrastructure mature.