The Ethics of Manuscript Authorship: Best Practices for Attribution
The International Committee of Medical Journal Editors (ICMJE), has established four criteria that each author of a paper should meet. This article and our free white paper, Credit Where Credit Is Due, detail and explore these criteria.
Updated on July 25, 2013
Authorship is becoming an increasingly complicated issue as research collaborations proliferate, the importance of citations for tenure and grants persists, and no consensus on a definition is reached. This issue is fraught with ethical implications because clearly conveying who is responsible for published work is integral to scientific integrity.
Many journals currently adhere to the guidelines of the International Committee of Medical Journal Editors (ICMJE), which has established four criteria that each author of a paper should meet:
- Significant involvement in study conception/design, data collection, or data analysis/interpretation;
- Involvement in drafting or revising manuscript;
- Approval of final version of manuscript for publication; and
- Responsibility for accuracy and integrity of all aspects of research.
Download our free white paper on authorship for a copy of these criteria and our suggestions for choosing authors appropriately.
Moreover, by the ICMJE definition, authors “should be able to identify which co-authors are responsible for specific other parts of the work...[and] have confidence in the integrity of the contributions of their co-authors.” Based on this description and the fourth criterion, authorship implies not only past individual contribution to a research project but also ongoing joint accountability for that project. As a result, authors may share fame or infamy, depending on the validity of the work.
The ICMJE also notes that an author must have made “substantive intellectual contributions” to the manuscript. Creative input is thus more eligible for authorship than purely mechanical work. A technician merely acquiring data, a senior researcher only obtaining funding or providing supervision, a collaborator solely providing a new reagent or samples, and other research-related but non-creative tasks do not merit authorship on their own. These individuals and their contributions could be cited in an acknowledgments section instead.
Despite this clearly outlined definition, numerous issues (including ethical concerns) have arisen regarding authorship attribution. These issues have emerged partly because many journals continue to adhere to their own guidelines or to various modified versions of the ICMJE criteria (see, for example, Table 2 in this EMBO reports article) and partly because the ICMJE guidelines may be insufficient, as argued at the 2012 International Workshop on Contributorship and Scholarly Attribution. A selection of topics that is specifically pertinent to academia is as follows:
The specific roles of individual authors in a research project are not always clear, especially when a manuscript is attributed to a large group. To address this problem, several journals (such as PNAS) require public disclosure of the specific contributions of each author. Some have also suggested the establishment of a database or the use of existing research community networks (such as ResearchGate) to track contributions. This tracking is particularly relevant because scholarly output is increasingly defined by metrics beyond paper citations (also known as altmetrics). To further clarify the roles of authors and encourage integrity, certain journals require a public guarantor for each article, or an author who takes responsibility for the entire research project, including conception, data acquisition and analysis, and publication. Ambiguity surrounding authorship may also arise from the publication of papers by researchers with the same name but could be minimized by the use of an ORCID identifier.
The meaning of the list order of authors on a paper varies between fields. In certain areas, the list is alphabetical, whereas in others, the convention includes citing every person who contributed in some way to the project (which may conflict with the ICMJE guidelines). In many disciplines, the author order indicates the magnitude of contribution, with the first author adding the most value and the last author representing the most senior, predominantly supervisory role. In this model, disputes may arise regarding who merits sole or shared first authorship. The Committee on Publication Ethics recommends that researchers discuss authorship order from project initiation to manuscript submission, revising as necessary, and record each decision in writing. Furthermore, contributions could be quantified, such as based on a points system (subscription required), to facilitate authorship decisions.
Honorary authorship is given to an individual despite a lack of substantial contributions to a research project. One form, gift authorship, is bestowed out of respect for or gratitude to an individual. For example, in Asian cultures, departmental heads or senior researchers may be added to a paper regardless of their involvement in the research. Another form, guest authorship, may be used for multiple purposes, including to increase the apparent quality of a paper by adding a well-known name or to conceal a paper's industry ties by including an academic author. Additional issues regarding honorary authorship are the inclusion of an author on a manuscript without his or her permission (which is often prevented by journal guidelines that require the consent of all authors) and coercive authorship, which typically consists of a senior researcher (such as a dissertation advisor) forcing a junior researcher (such as a graduate student) to include a gift or guest author.
Honorary authorship is a major ethical issue in scholarly publication, as this dishonest practice was found in approximately 18% of articles in six medical journals in 2008. From the standpoint of journals, lists of specific contributions may help to minimize this practice, as could reminders that all authors are accountable for the integrity of a published work. The institution of double-blind peer review could also decrease the influence of authors' prominence in the field on journal acceptance. At research institutions, guidelines could equate honorary authorship with research misconduct. Additionally, the donation of resources to a project without the expectation of automatic authorship could be encouraged by the use of contributions, including those listed in acknowledgments sections, as a measure of output, as discussed above.
In all cases described here, more universal standards for manuscript authorship will be critical for fostering good practices. As you write and review manuscripts, remember the best practices found in this white paper, and consider ways to bring authorship credit and accountability to the attention of your colleagues and readers.