Beyond authorship: Attribution, contribution, collaboration, and credit

Abstract

• As the number of authors on scientific publications increases,
ordered lists of author names are proving inadequate for the
purposes of attribution and credit.
• A multi-stakeholder group has produced a contributor role taxonomy
for use in scientific publications.
• Identifying specific contributions to published research will lead
to appropriate credit, fewer author disputes, and fewer disincentives
to collaboration and the sharing of data and code.

Full text

Beyond authorship: attribution, contribution, collaboration, and credit 151
LEARNED PUBLISHING VOL. 28 NO. 2 APRIL 2015
Beyond authorship: attribution, contribution, collaboration, and credit
Amy Brand et al.
Learned Publishing, 28: 151–155
doi:10.1087/20150211
INDUSTRY UPDATE
Beyond authorship:
attribution,
contribution,
collaboration, and
credit
Amy Brand Digital Science
Liz Allen Wellcome Trust
Micah Altman MIT Libraries
Marjorie Hlava Access Innovations
Jo Scott Wellcome Trust
Key points
• As the number of authors on scientic publications increases,
ordered lists of author names are proving inadequate for the
purposes of attribution and credit.
• A multi-stakeholder group has produced a contributor role tax-
onomy for use in scientic publications.
• Identifying specic contributions to published research will lead
to appropriate credit, fewer author disputes, and fewer disincen-
tives to collaboration and the sharing of data and code.
Most researchers who have co-authored articles
for publication, whether with one collaborator
or twenty, have a story about wrangling over
the order in which author names appear in the byline.
And every journal publisher, large or small, deals regu-
larly with cases of author dispute.1
In the 1930s, the average number of collaborators
on scientic papers was roughly two, and this number
remained steady for four decades.2 Authorship and col-
laboration have changed dramatically since the 1970s,
and growth in multi-authorship has accelerated, driven
both by academic reward systems and the ease of collab-
oration in the Internet age. By 2000, the average number
of authors in articles published in high-ranking medical
journals was seven. Before 1975, the maximum number
of authors associated with any article in MEDLINE was
38,3 whereas it is not unusual today for scientic publi-
cations to list hundreds or thousands of authors.4 At the
same time, interdisciplinary collaboration has increased,
and other forms of scholarly output, including data and
software, are now published in citable form.5 Some of
these new scholarly collaborations, in particular citizen-
science projects such as the Sloan Galaxy Zoo, can
attract hundreds of thousands of named contributors.6
As the average number of authors on scientic arti-
cles grows, authorship-related problems, ranging from
disputes to outright misconduct, mount. Why, then, do
we persist with a practice of attributing scientic con-
tribution that fails to capture the true nature of the
underlying collaboration – or, more precisely, to cap-
ture who did what? It’s not as though the stakes here
Amy Brand Liz Allen Micah Altman Marjorie Hlava Jo Scott
© Amy Brand, Liz Allen, Micah Altman, Marjorie Hlava, and Jo Scott 2015

152 Beyond authorship: attribution, contribution, collaboration, and credit
LEARNED PUBLISHING VOL. 28 NO. 2 APRIL 2015
are inconsequential. Who gets credit for discovery or
creation has a tremendous impact on people’s lives. It
affects career advancement and tenure in the academic
sphere, and the transparency and integrity of the perma-
nent research record.
When there are multiple authors, we tend to rely
on the order in which names are listed to infer lead
contribution, but in fact there are no consistent name
ordering practices from one eld to the next.7,8 Even
weak conventions around ordering break down in mul-
tidisciplinary collaborations when eld-specic practices
conict. In elds such as economics, in which the order
of names defaults to alphabetical, and it is typically
assumed that all authors contribute equally, it has even
been shown that you are somewhat more likely to get
tenure or win a prestigious prize if your last name begins
with a letter earlier in the alphabet.9
In elds without the alphabetical order convention,
decisions about lead authorship can be especially con-
tentious. How we apportion credit for collaborative
works today is highly subjective, open to abuse, and
often determined more by laboratory politics or senior-
ity than by actual effort or contribution.10 In these
situations, junior researchers, for whom the reputational
stakes are especially high, and those making non-tradi-
tional research contributions, such as in the form of data
or software code, tend to lose out most on the recogni-
tion they deserve.
A separate but related question concerns what
qualies an individual contributor for authorship sta-
tus, and this, too, is often contested. Within the
biomedical community, the authorship guidelines pro-
duced by broadly recognized organizations such as the
International Committee of Medical Journal Editors
(ICMJE) and the Committee on Publication Ethics
(COPE) focus primarily on who should be listed as
author, and what processes should be used for authorship
disputes and corrections. Within the eld of medi-
cine, conventions vary across publication venues and
institutions.11 For example, Harvard Medical School’s
authorship rules specify that, ‘Everyone who has made
substantial intellectual contributions to the work should
be an author’ (see http://hms.harvard.edu/about-hms/
integrity-academic-medicine/hms-policy/faculty-policies-
integrity-science/authorship-guidelines). Similarly, the
Proceedings of the National Academy of Sciences (PNAS)
require that ‘authorship must be limited to those who
have contributed substantially to the work’ (see http://
www.pnas.org/site/misc/iforc.pdf).
The ICMJE policy, on the other hand, limits author-
ship to those who make
substantial contributions to the conception or design
of the work; or the acquisition, analysis, or interpre-
tation of data for the work; and drafting the work or
revising it critically for important intellectual content;
and nal approval of the version to be published; and
agreement to be accountable for all aspects of the work
in ensuring that questions related to the accuracy or
integrity of any part of the work are appropriately
investigated and resolved. (See http://www.icmje.org/
recommendations/browse/roles-and-responsibilities/
defining-the-role-of-authors-and-contributors.html
emphasis added)
Awareness of problems with the conventional author-
ship model is by no means new. The topic received a
great deal of attention in the late 1990s, for example, in
the context of accountability in medical journal publish-
ing. The work of Drummond Rennie on this subject was
particularly inuential.12 In a 1997 article in the Journal
of the American Medical Association (JAMA), Rennie and
co-authors wrote:
The system of authorship, while appropriate for
articles with only one author, has become inappro-
priate as the average number of authors of an article
has increased; as the work of coauthors has become
more specialized and relationships between them
have become more complex; and as both credit and,
even more, responsibility have become obscured and
diluted. Credit and accountability cannot be assessed
unless the contributions of those named as authors
are disclosed to readers, so the system is awed. We
argue for a radical conceptual and systematic change,
to reect the realities of multiple authorship and to
buttress accountability. We propose dropping the out-
moded notion of author in favor of the more useful
and realistic one of contributor. This requires dis-
closure to readers of the contributions made to the
research and to the manuscript by the contributors,
so that they can accept both credit and responsibility.
Whether from the perspective of credit or accountability,
clearly we need a better system for representing collab-
orative contribution to published works – something
more akin to lm credits.13 In the intervening years since
Rennie’s radical call to action, several medical and life
science publishers have started to collect contribution
statements for multi-authored works. Some publishers,
such as the American Association for Cancer Research
(AACR) and the Public Library of Science (PLOS), ask
authors to select roles from a predened list. Others,
such as Nature, invite or require free-text contribution
statements, yet many publishers who collect role infor-
mation from authors do not even publish it.
What the scholarly publishing community still lacks
is coordination among contributorship efforts. In the
absence of standardization and coordination, the infor-

Beyond authorship: attribution, contribution, collaboration, and credit 153
LEARNED PUBLISHING VOL. 28 NO. 2 APRIL 2015
Table 1. CRediT – contributor role taxonomy
Header: This taxonomy provides a high-level classication of the diverse roles performed in the work leading to a published research
output in the sciences. Its purpose is to provide transparency in contributions to scholarly published work, to enable improved
systems of attribution, credit, and accountability.
The classication includes, but is not limited to, traditional authorship roles. That is, these roles are not intended to dene what
constitutes authorship. Rather, the roles are intended to apply to all those who contribute to research that results in scholarly
published works, and it is recommended that all tagged contributors be listed, whether they are formally listed as authors or named in
acknowledgements.
An individual contributor may be assigned multiple roles, and a given role may be assigned to multiple contributors. When there are
multiple people serving in the same role, a degree of contribution may optionally be specied as ‘lead’, ‘equal’, or ‘supporting’. It is
recommended that corresponding authors assume responsibility for role assignment, and that all contributors be given the opportunity
to review and conrm assigned roles.
Term Denition
Conceptualization Ideas; formulation or evolution of overarching research goals and aims
Methodology Development or design of methodology; creation of models
Software Programming, software development; designing computer programs; implementation of the
computer code and supporting algorithms; testing of existing code components
Validation Verication, whether as a part of the activity or separate, of the overall replication/
reproducibility of results/experiments and other research outputs
Formal Analysis Application of statistical, mathematical, computational, or other formal techniques to
analyze or synthesize study data
Investigation Conducting a research and investigation process, specically performing the experiments,
or data/evidence collection
Resources Provision of study materials, reagents, materials, patients, laboratory samples, animals,
instrumentation, computing resources, or other analysis tools
Data curation Management activities to annotate (produce metadata), scrub data and maintain research
data (including software code, where it is necessary for interpreting the data itself) for
initial use and later reuse
Writing – Original Draft Preparation, creation and/or presentation of the published work, specically writing the
initial draft (including substantive translation)
Writing – Review & Editing Preparation, creation and/or presentation of the published work by those from the original
research group, specically critical review, commentary or revision – including pre- or post-
publication stages
Visualization Preparation, creation and/or presentation of the published work, specically visualization/
data presentation
Supervision Oversight and leadership responsibility for the research activity planning and execution,
including mentorship external to the core team
Project Administration Management and coordination responsibility for the research activity planning and
execution
Funding acquisition Acquisition of the nancial support for the project leading to this publication.

154 Beyond authorship: attribution, contribution, collaboration, and credit
LEARNED PUBLISHING VOL. 28 NO. 2 APRIL 2015
mation about contribution that publishers may already
be collecting is not making its way into the metadata
network that underpins our citation and credit systems.
What we need is a controlled vocabulary of contribu-
tor roles and mechanisms for capturing contribution tags
within the scholarly metadata ecosystem.
Imagine publishers collecting structured information
about contribution in a standard format. Imagine, fur-
ther, that this information is associated with the article
DOI, via CrossRef, and with ORCID author identiers.
We would then have the infrastructure in place to track
not only who authored which publications, but also
who contributed what to each publication that names
the individual as a contributor. With this infrastructure
in place, it would eventually be possible to devise more
precise, author-centric credit and impact tracking tools,
on which the byline order of author names would have
no bearing.
In May 2012, we hosted a workshop at Harvard
University to explore this topic with representatives of
the publishing, funding, and academic worlds.14 A key
outcome of this workshop was the commitment by a
sub-group of attendees to devise a high-level contributor
role taxonomy for the sciences. We drafted a preliminary
taxonomy by analyzing acknowledgments and free-text
contribution statements, and conducted a survey study
in partnership with several publishers to gauge the feasi-
bility of asking corresponding authors to assign the roles.
The results of the study were overwhelmingly positive.
These efforts are described in a Nature commentary arti-
cle published last year.15
Based on the success of the 2013 study, we are
partnering with two information industry standards
organizations, the Consortia Advancing Standards
in Research Administration Information (CASRAI)
and the US-based National Information Standards
Organization (NISO), to achieve broader community
consultation in rening the taxonomy and testing its t
with a range of scientic elds. During the latter half of
2014, a 17-person working group composed of represen-
tatives from several publishers, funders, and universities
met monthly under the auspices of CASRAI to review
and rene each of the roles and role descriptions. This
effort adopted the name Project CRediT, and the proj-
ect overview and the taxonomy itself are available at
http://projectcredit.net. Once we reached consensus on
the 14-term taxonomy – see Table 1 – we opened the
project up for public comment and received over 100
responses using an online feedback form. Researchers
constituted 75% of the respondents, and the group was
fairly diverse in terms of geographic make-up, with the
biological sciences being more strongly represented than
other researcher areas.
As with the earlier study, the results of the feedback
process for this version of the taxonomy were highly
encouraging. A clear majority of respondents agreed
with all of the proposed terms. Most of the questions
that arose concerned confusion over whether the tax-
onomy was explicitly intended to specify which types of
contribution qualify for authorship status, when in fact
that was never the intention. As stated in the taxonomy
header:
The classication includes, but is not limited to,
traditional authorship roles. That is, these roles are
not intended to dene what constitutes authorship.
Rather, the roles are intended to apply to all those who
contribute to research that results in scholarly pub-
lished works, and it is recommended that all tagged
contributors be listed, whether they are formally
listed as authors or named in acknowledgements.
Among the other recommendations to emerge from the
public consultation process were: (1) to adopt a coarse-
grained degree of contribution, as an optional tag to be
used in conjunction with a contributor role when more
than one contributor serves in the same role; and (2)
to have corresponding authors be responsible for role
assignment, but only with review and conrmation by
all contributors. The taxonomy header captures both of
these recommendations as follows:
An individual contributor may be assigned multiple
roles, and a given role may be assigned to multiple
contributors. When there are multiple people serv-
ing in the same role, a degree of contribution may
optionally be specied as ‘lead’, ‘equal’, or ‘support-
ing’. It is recommended that corresponding authors
assume responsibility for role assignment, and that all
contributors be given the opportunity to review and
conrm assigned roles.
We are currently working with several publishing
partners and providers of manuscript-tracking and
author-submission systems to undertake early implemen-
tations of the taxonomy. We expect a number of these
implementations to be up and running later in 2015. An
early pilot of the taxonomy by Mozilla Science Labs is
already underway. It uses the taxonomy in a set of digital
contributorship badges maintained at the browser level,
as described in a recent Science Magazine news piece.16
As word about the Project CRediT taxonomy has
spread through ongoing conference presentations and
coverage in leading community blogs,17 new efforts
are focused on implementation pathways, including
integrating the taxonomy into the National Library
of Medicine’s Journal Article Tag Suite (JATS) DTD
(http://jats.nlm.nih.gov/about.html). The Force11 com-

Beyond authorship: attribution, contribution, collaboration, and credit 155
LEARNED PUBLISHING VOL. 28 NO. 2 APRIL 2015
munity (https://www.force11.org/about) has created a
new working group to look at attribution implementa-
tion for all research products.
With these aligned efforts, and the groundswell of
interest among researchers, funding agencies, academic
institutions, and editors in increasing the transparency
of research contributions, standardized contribution
tagging, while still early-stage, is gaining rm footing in
scholarly journal publishing. If this initiative is ultimately
successful, there will be far fewer author disputes, and
fewer disincentives to collaboration and the sharing of
data and code, for example, because those contributions
will be more reliably recognized. Among the less obvious
benets is enhanced mineable information on research
expertise, for the purposes of research networking and
peer-reviewer identication.18 Hence these efforts could
positively inuence both the cooperative culture of
research, and academic incentive structures more gen-
erally. We invite authors and publishers alike to follow
the example below in describing contribution using the
CRediT taxonomy.
Author statement and acknowledgements
The rst author named is lead and corresponding author. All other authors
are listed in alphabetical order. We describe contributions to the paper
using the taxonomy provided above. Writing – Original Draft: A.B. and
M.A.; Writing – Review & Editing: M.A., A.B., and M.H.; Conceptualization:
L.A. and A.B.; Investigation: L.A., A.B., M.A., and M.H.; Methodology:
M.A. and J.S.; Formal Analysis: M.A. and J.S.; Project Administration: L.A.
and A.B.; Funding Acquisition: L.A. and A.B.
The work described in this article was supported by the Wellcome Trust
and Digital Science.
The authors would like to acknowledge the other members of the
CASRAI working group who provided critical review of the taxonomy,
but are not responsible for the content of this article: Helen Atkins, David
Baker, Monica Bradford, Todd Carpenter, Jon Corsant-Rikert, Jeffrey
Doyle, Melissa Haendel, Daniel S. Katz, Veronique Kiemer, Nettie Lagace,
Emile Marcus, Walter Schaeffer, Gene Sprouse, and Victoria Stodden.
References
1. Levsky, M.E, Rosin, A., Coon, T.P., Enslow W.L., and Miller, M.A.
2007. A descriptive analysis of authorship within medical journals,
1995–2005. Southern Medical Journal, 4(1): 371–375. http://dx.doi.
org/10.1097/01.smj.0000257537.51929.4b
2. Clarke, B.L. 1964. Multiple authorship trends in scientic papers.
Science, 143(3608): 822–824. http://dx.doi.org/10.1126/science.143.
3608.822
3. Number of Authors per MEDLINE®/PubMed® Citation, http://
www.nlm.nih.gov/bsd/authors1.html
4. King, C. 2012. Multiauthor papers: onward and upward. ScienceWatch
Newsletter. http://archive.sciencewatch.com/newsletter/2012/201207/
multiauthor_papers/
5. Altman, M. and Crosas, M. 2013. The evolution of data citation:
from principles to implementation. IASSIST Quarterly, 37: 62.
6. Raddick, M.J., Georgia, B., Pamela, L.G., Lintott, C.J., Murray, P.,
Schawinski, K., Szalay, A.S., and Jan Vandenberg, J. 2010. Galaxy
zoo: exploring the motivations of citizen science volunteers.
Astronomy Education Review, 9(1): 010103. http://dx.doi.org/10.3847/
AER2009036
7. Academic authorship, 22 January 2015. In Wikipedia, The Free
Encyclopedia. Retrieved 15.47, 7 February 2015, from http://en.wiki-
pedia.org/wiki/Academic_authorship
8. Wager, E. 2007. Do medical journals provide clear and consistent
guidelines on authorship? Medscape General Medicine, 9(3): 16.
9. Einav, L. and Yariv, L. 2006. What’s in a surname? the effects of sur-
name initials on academic success. Journal of Economic Perspectives,
20(1): 175–187. http://dx.doi.org/10.1257/089533006776526085
10. Dance, A. 2012. Who’s on rst? Nature, 489: 591–593. http://dx.doi.
org/10.1038/nj7417-591a
11. Bosnjak, L. and Marusic, A. 2012. Prescribed practices of authorship:
review of codes of ethics from professional bodies and journal guide-
lines across disciplines. Scientometrics, 93(3): 751–763.
12. Rennie, D., Yank, V., and Emanuel, L. 1997. When authorship fails. A
proposal to make contributors accountable. JAMA, 278(7): 579–585.
http://dx.doi.org/10.1001/jama.1997.03550070071041
13. Patterson, M. 2007. Roll credits: sometimes the authorship byline
isn’t enough. Guest PLoS Blog by Michael Molla and Tim Gardner.
Posted 6 November 2007. http://blogs.plos.org/plos/2007/11/
roll-credits-sometimes-the-authorship-byline-isnt-enough/
14. IWCSA Report 2012. Report on the International Workshop on
Contributorship and Scholarly Attribution, 16 May 2012. Harvard
University and the Wellcome Trust. http://projects.iq.harvard.edu/
attribution_workshop.
15. Allen, L., Scott, J., Brand, A., Hlava, M., and Altman, M. 2014.
Credit where credit is due. Nature, 508: 312–313. http://dx.doi.
org/10.1038/508312a
16. Dalmeet, S.C. 2014. Could digital badges clarify the roles
of co-authors. Science, published online 3 November 2014.
http://news.science mag.org/ scientific- community/ 2014/ 11/
could- digital- badges- clarify-roles- co- authors
17. Meadows, A. 2014. An interview with Amy Brand on a proposed
new contributor taxonomy initiative. Scholarly Kitchen, published
online 20 August 2014. http://scholarlykitchen.sspnet.org/2014/08/20/
an-interview-with-amy-brand-on-a-proposed-new-contributor-taxon-
omy-initiative/
18. Academy of Medical Sciences. 2015. Team Science Working Group
– project update February 2015. http://www.acmedsci.ac.uk/down-
load.php?f=le&i=30834
Amy Brand (corresponding author)
Vice President
Digital Science
25 First Street Suite 104
Cambridge MA 02141, USA
a.brand@digital-science.com
Liz Allen
Wellcome Trust
Micah Altman
MIT Libraries
Marjorie Hlava
Access Innovations
Jo Scott
Wellcome Trust