10.2 Guidelines for Authorship

Publication practices vary widely between disciplines, so that a single set of firm policies governing authorship is not possible or desirable. The following guidelines are intended to prompt discussion of expectations at the department level. The recommended best practice is for each department to develop its own special guidelines in order to inform graduate students, academic staff, and other collaborators in research.

Publication of thesis work and other research results is important for both the author(s) and the institution where the work was conducted. Authorship should be reserved for those and only those who have made significant intellectual contributions to the research. For manuscripts derived from thesis work, the student normally would be the first or senior author (if there were multiple individuals contributing to the work). Students who are involved in research other than thesis research should be involved as authors on resulting publications if their contributions warrant, according to the criteria outlined below. To avoid misunderstandings, authorship should be discussed and agreed upon well in advance of a manuscript being completed.

Over the years, various standards or criteria for authorship have been proposed and defined. Many journals and journal editors now accept the basic criteria defined by Ed Huth (1986 Ann Int Med 104:269-274):

  1. All authors should have made a substantial contribution to the conception, design, analysis, or interpretation of data;
  2. they should have been involved in writing and revising the manuscript for intellectual content; and
  3. they should have approved the final draft and be able to defend the published paper.

Those who have made other contributions to the work (such as data collection without interpretation) or only parts of the above criteria should be credited in the acknowledgements, but not receive authorship.

Since the order of authorship still varies somewhat by discipline, the primary (often referred to as senior) author should always be identified as such. In most fields, the primary or senior author is listed first, with the other authors listed in descending order of importance. In other journals, the order of authors after the primary author varies. In some, the author who made the next greatest contribution (often the student's supervisor) is listed last. In a few fields, the order of authorship is determined by chance (flip of a coin) or by alphabetical order. To avoid any confusion over the roles of each author on multi-authored papers, some journals require that the specific role of each author in the research be explicitly spelled out in a footnote. This practice is admirable and would resolve many of the difficulties and disagreements regarding authorship.

For collaborative projects where the student and supervisor have agreed to jointly publish their work, a department should set some time limit (with possibly a 1-year limit as a default) after completion of the thesis for the student to complete and submit a manuscript to a journal, unless otherwise agreed to and recorded by the student and faculty member. If the student does not meet a reasonable time limit and there appears to be no progress towards publication, the supervisor of such a collaboration would have the right to complete and submit a manuscript with the student as an author.

Institutional affiliations of an author should normally be identified with the place where the author was employed or under whose auspices most of the work was carried out. In special cases the University may withhold the use of its name as an affiliation where it believes that the work is not of an acceptable standard.

Footnotes should be used for current addresses, if different from the place where the work was carried out. For example, if a PhD student at the U of A publishes a paper based upon MSc work at McGill, the address byline should carry the McGill address with a footnote to the author's name stating, for example, "current address: Department of Biological Sciences, University of Alberta, Edmonton, Alberta, Canada." Similarly, the same student, now a postdoctoral fellow at Harvard, might publish a paper based upon the PhD work at the University of Alberta; in that case the address byline on the paper would be the U of A address, with a footnote to the name giving the Harvard address.

Authorship also involves other ethical responsibilities, some of which are outlined in the University of Alberta Research and Scholarship Integrity Policy. In addition, authors should declare any potential conflicts of interest to the journal editor when they submit a manuscript. Obvious situations include: 1) receiving support from a tobacco company for work on the health benefits of smoking; 2) owning stock in a drug company when reporting on effects of one of the company's drugs.

(FGSR Council, 1996/11/15)