Author: Michael Kalichman, 2001
Contributors: P.D. Magnus, Dena Plemmons
- Credit and responsibility
While the list of authors identifies those who deserve credit for the work being published, those authors also bear responsibility for any deficits in the integrity or quality of the work.
- Who should be an author?
Because authorship is a matter of public credit and responsibility, those and only those who have met accepted criteria for authorship should be included as authors.
Research groups and collaborators should be clear about the criteria and plans for authorship; individual scientists should discuss authorship during the planning of any collaboration and continue those discussions as the research project evolves.
Authorship is the most visible form of academic recognition and credit. However, because credit for publication is also important in disputes and allegations of research misconduct, it is worth considering why authorship credit is more than a matter of personal gratification. Indeed, attribution of credit and responsibility is central to the structure of science.
The framework of science depends in part on the ability of institutions, policy makers, and the public to identify who is responsible for the work and its interpretation. Funding agencies consider past success, as evidenced by authorship, in the allocation of research grants. Research institutions often use authorship as evidence of creative contributions that warrant promotion. Scientists themselves may use credit for past work as a mechanism to attract both new trainees and willing collaborators. Finally, in an era of increasing emphasis on commercialization, authorship and credit help to define intellectual property rights. These and other reasons explain scientists' desire for the credit of authorship, and also make clear why the assignment of authorship is central to the responsible conduct of research.
Despite the importance of authorship credit, nearly all aspects of authorship and publication are covered only by guidelines and unspoken custom. One consequence of this is that authorship practices can vary dramatically among disciplines and institutions, and often between labs and departments in the same discipline and institution.
One definition of authorship accepted by many medical journals is that adopted by the International Committee of Medical Journal Editors (ICMJE) . Under this definition, someone is an author if and only if they have done all of the following:
- made substantial contributions to conception and design, or acquisition of data, or analysis and interpretation of data;
- drafted the article or revised it critically for important intellectual content; and
- approved of the final version to be published.
In recent years, a new model of authorship was proposed by an Authorship Task Force of the Council of Biology Editors (now the Council of Science Editors). This model is now also endorsed by the ICMJE (2006). For the community of scientists, transparency about authorship contributions is accomplished simply by publishing the way in which individual authors contributed to the work. The 'contributorship' model is less restrictive than the ICMJE model in defining authorship, but the contributions of each author are identified to the journal and published with the manuscript (Horton and Smith, 1996; Smith, 1997; Rennie et al., 2000; Authorship Task Force, 2000). Several medical journals now use this model.
- List and describe advantages of authorship. Are there circumstances under which it would be disadvantageous to be an author? If so, why?
- When and how have the criteria for authorship been discussed in your research group? What are the criteria? If this is not clear, then what steps could you take to better define the criteria for yourself and others?
- List and describe responsibilities of authorship.
- Describe the ICMJE guidelines and the contributorship model for authorship. What are the advantages and disadvantages to these two different approaches?
Authorship might be justified by significant contributions to the ideas that preceded the work, design of the study, execution of the study, data analysis, or drafting of the manuscript. Yet some questions about who deserves authorship are not easily answered. Can simply performing the data collection ever be enough to justify authorship? Should it be necessary that every author be able to defend all aspects of a manuscript, or only some? Correspondingly, should all authors bear equal responsibility if any part of a manuscript is later found to depend on falsified or fabricated data?
- Credit: Institutions, funding agencies, and researchers assess scientists in light of their publications. Thus, including someone among the list of authors for a publication is taken to mean that they deserve credit for that publication.
- Responsibility: Credit for authorship is highly valued, but researchers sometimes forget that the privilege of authorship also comes with responsibility. If the work is later found to be irresponsible or misrepresented, then all authors will be associated with the work. Thus, all authors share responsibility for assuring that the studies and findings have been represented truthfully.
- Variable criteria: Methods of assigning authorship vary greatly in academia, even within the same institution or discipline. While it is widely agreed that authorship should be based on a substantial contribution, reasonable people can differ considerably over the definitions of both 'substantial' and 'contribution.' Some emphasize the importance of having done the work as a criterion, or the only criterion, for authorship. Others put more emphasis on ideas, experimental design, and data interpretation. In some research groups, decisions about authorship are made solely at the discretion of the principal investigator, while in other groups, decisions are made collectively by all who have had a significant role in the project. Some investigators expect authorship in return for providing access to key equipment, samples of an unusual reagent or cell line, or assistance with statistical methods or experimental design. Others argue that these contributions warrant only an acknowledgment, not authorship. However authorship is determined for a particular group, the methods of assigning authorship should be communicated early and often, and with a commitment to transparency.
- Minimal criteria:
Although criteria for authorship vary, an author ought at least minimally to have:
- made a substantial and new contribution to the research
- agreed to take responsibility for at least some of the content of the manuscript, including a review of the relevant raw data
- read and agreed to the manuscript before publication, and agreed to be named as an author
- Acknowledgment: Many elements essential for a publication should be credited, but do not warrant authorship. People who provide facilities or resources, for instance, should be credited in the Acknowledgments section. Authors have the ethical responsibility to acknowledge those who made the research and manuscript possible. Because agreement with the contents of a manuscript might be inferred, it is good practice, and sometimes required, that anyone who is acknowledged has given his or her permission to be listed.
- Authorship Task Force (2000): Is it time to update the tradition of authorship in scientific publications? Council of Science Editors (formerly Council of Biology Editors)
- Horton R, Smith R (1996): Signing up for authorship. Lancet 347(9004):780.
- International Committee of Medical Journal Editors (2006): Uniform Requirements for Manuscripts Submitted to Biomedical Journals. JAMA 4277:927-34
- Rennie D, Flanagin A, Yank V (2000): The contributions of authors. JAMA 284(1): 89-91.
- Smith R (1997): Authorship is dying: long live contributorship. British Medical Journal 315(7110): 696.
- Yank V, Rennie D (1999): Disclosure of researcher contributions: a study of original research articles in The Lancet. Annals of Internal Medicine 130(8): 661-70.