blog

Original research articles with one author – particularly in the life sciences – are increasingly rare, and the concept of ‘authorship’ in science has become outdated. Adopting a simple taxonomy of terms to describe the contributions to a published work could enable a range of benefits to all the stakeholders in research – most particularly allowing researchers a new basis upon which to have their contributions recognised.

Among the reasons why change is timely:

(i) There are many contributors to research; reliance on author position to decipher either level or nature of contribution to a published scientific article is increasingly inaccurate (was it ever accurate?), as big, collaborative and ‘team’ science becomes more commonplace in many areas of science.

(ii) Metrics based around individual publication output and contributions are relied upon to help further careers; if we can bring greater transparency to individual contributions to scholarly publication, then that can only be a positive thing – and particularly for early career researchers building a career.

(iii) The San Francisco Declaration on Research Assessment, endorsed by many researchers, research funders, learned societies and publishers, emphasized a commitment to move away from Journal Impact Factor as a measure of research quality. Initiatives that bring greater clarity to the myriad contributions to science would provide a new basis upon which to recognise researcher contributions, and to consider use and re-use of particular contributions (e.g. data; software).

(iv) Technology supporting the publication of research can now facilitate capture of meta-data (including contributor roles) across a wider range of fields than ever before.
There are also practical uses in knowing who contributed to specific components of published work, such as helping funders to identify potential grant peer reviewers with a defined specialism (e.g. statistician; data curator), and helping researchers to identify and forge collaborations with individuals with specific skills. For publishers, of course, there are obvious benefits to greater transparency in contributor assignment as it can reduce the volume and time involved in managing authorship disputes.

Testing the contributor role taxonomy in practice

A working group comprising researchers, research institutions, funding agencies, publishers and learned societies, developed a simple taxonomy of 14 roles that could be used to assign contribution types to scholarly published outputs – particularly in the life sciences (see Credit where credit’s due for project background). The taxonomy recognizes roles like data curation, development of design methodology, programming and software development, application of statistical or mathematical techniques to analyze data, and data visualization.

The standards organisation, CASRAI (Consortia Advancing Standards in Research Administration), is now the custodian of the taxonomy, and are capturing usage feedback and providing guidance on how to implement the roles in various scholarly workflows (Project CRediT). Several organisations are already testing the taxonomy of contributor roles in a live environment: Cell Press have endorsed the use of the roles amongst their ‘authors’; Aries Systems has included the taxonomy in its recent release of the Editorial Manager 13.0 manuscript submission system; the Mozilla Science Labs Open Badges project has included the roles in its prototype; and ORCID plan to incorporate the contributor roles into their Registry to enable specific contributions to research works (such as an article, dataset or presentation) to be displayed on an ORCID record. Other publishers and platform providers, including F1000, are also exploring how to shift their existing model of capturing contributions to a more systematic and transparent approach, which the CRediT taxonomy enables.

As project Co-Chair, a role I share with Amy Brand currently Head of MIT Press, when overhauling an established way of working, you have to be careful not to bring about unhelpful or negative consequences. However, without trying to evolve and upgrade what is now a rigid and, in many ways, unhelpful model of authorship, we will not know if things can be better for science.

Liz Allen

Director of Strategic Initiatives at F1000 & Visiting Senior Research Fellow at Kings’ College London. Liz is also a current Co-Chair of the Project CRediT working group