CORRESPONDENCE ON AN ETHICAL ISSUE
CONCERNING AUTHORSHIP CREDIT

Date: Fri, 12 Nov 1999 12:31:59 -0800
Subject: Ethics of Authorship

Colleagues,

I would appreciate your perspectives on the following situation in the training program I direct:

1. A student completes MA thesis. The idea for the thesis is provided to the student and a very substantial amount of guidance in the development of the design, approach to analysis, and writing is given to the student. The advisors views his/her contribution to be more than ordinary and the progress to be slower than ordinary. 2. The student demures when a timetable for revision for publication is proposed, abd indicates no interest in writing it up in the near future. The student goes on leave, working on other projects in the university. 3, The project is given to another member of the advisor's research group to work on. That person found some errors in the data base and re-ran the analyses. There were no substantive changes in results. That person also identified some interesting alternative ways to look at the data, additional literature and wrote the drafts, on which the advisor collaborated. 4. The student has seen the last draft on which relatively minor comments are made.

The issue of authorship arises. The advisor believes that the principal intellectual contributions are the advisor's and the new contributor 's, that the ms. would not ever see the light of day in a reasonable time without the leadership of the new person, and so the student should not be first author. The student believes that he should be because of all of the work that was done.

Suggestions? Opinions? Thanks!


Date: Fri, 12 Nov 1999 12:50:21 -0500

I guess that if this were my student, I would still list him as first author on the article. The only time I made an exception to this was when a study based on a master's thesis was combined with a replication and extension of it by a third party into a single manuscript.
 


Date: Fri, 12 Nov 1999 12:14:07 -0600

This is an interesting scenario and one that comes up quite frequently. My view is that an MA thesis student (in this scenario, student #1) should be given fair warning that if he/she is not prepared to write up the project for publication in a timely manner, the advisor (who in this case was principally responsible for the genesis of the project) or some other member of the group would take the lead (and first authorship) on the paper. Assuming student #1 in the present scenario was informed of this, my call would be for that student to be 2nd or possibly even third author.


Date: Fri, 12 Nov 1999 13:21:35 -0500

I have done what Don suggested over the years but recently while expressing my frustration with just such a situation as John described a very valued advisor of my, my wife, asked me a question that bears on the situation. "What will you be teaching your student if they are recognized as first author?" I would like to hear the answer of others to that question.
 


Date: Fri, 12 Nov 1999 14:10:56 -0500

There is an interesting and highly relevant discussion of the ethics of authorship the December 1998 issue of the APA Monitor. I recommend it highly. Perhaps I liked the article because it reinforced my own opinion that blind acceptance of the notion that papers based on a master's thesis (or even a dissertation) must ALWAYS list the student as first author. I do not agree with that assertion, in that it seems that the most ethical act is to list authors according to the level of contribution. Although in the best of all worlds the student would be the primary contributor, in many cases that is not the case.
 


Date: Fri, 12 Nov 1999 14:02:05 -0600

I discuss such issues with our first-year graduate students in the 2nd half of our 2-semester Clinical Research Seminar. We consider the APA Principles, and the article by Fine & Kurdek (Amer Psychologist, 1993, 48, 1141-1147) as well as comments in the December 1994 AP. Also, we look at a very useful article by Roger B. Winston, Jr. (1985). A suggested procedure for determining order of authorship in research publications. Journal of Counseling and Development, 63, 515-518.

Winston suggests that points be distributed among prospective authors for various categories of activity which differ in their relative importance. For example, conceptualizing and refining the research idea would be worth 50 points. In Milt's situation, presumably the faculty member would be assigned 40 and Student A gets 10 points (Student B is not yet involved so he/she gets 0). The literature search is valued at 20 points; presumably Student A had guidance here too but maybe we can divide the points equally, realizing that Student B later found additional literature. Let's give 8 to faculty member, 8 to A and 4 to B.
 

Creating the research design gets 30 pts. Instrument selection gets 10 pts. Instrument construction/ questionnaire design gets 40 pts. Assuming that these were all all involved and divided 50-50 between Faculty and A (and this is probably too kind to Student A), the faculty member's running total is 88, Student A's is 58, and Student B's is 4 points.
 

Data collection is valued at 10 points; these go to Student A.
 

Next the selection of statistical tests and analyses, performing statistical analyses and interpreting of statistical analyses each receive 10 pts. Here Student B has had a greater role and apparently A's work has been superceded. Let's assume that A gets 10 pts for the first task, B gets 10 for the second task, and faculty member gets 10 for the interpretation. (Again, these numbers may too fair to Student A although A may quibble with my point assignment too.)
 

Writing the first draft is worth 50 points; writing the second is worth 30. Editing is worth 10. Assuming that B did the major writing (with some initial input from Faculty Member on outline--say 5 points worth--and the Faculty member did the editing, we now have the following totals:
 

Faculty member: 113

Student A: 78

Student B: 89
 

The order of authorship should be Faculty, B, A. However, the relative contributions of the students are closer than might have been assumed.
 

Obviously, any conclusions could change with more than slight adjustments in the point assignments. Ideally these could be discussed by the people involved. This doesn't solve the basic problem that people tend to overestimate their own contributions to many tasks relative to what they perceived others to have done. But at least it's a start to their discussion. Perhaps on the basis of this kind of analysis, the faculty member (if he/she has already received tenure) might wish to negotiate an authorship order of B, A, Faculty given the relative contributions of the two students.
 

My personal strategy is to be very clear about authorship with masters students when I give them a problem, particularly if it's the next study I would wish to conduct in my own program; and to note the likelihood of changes as we often attempt to do multiple studies before publishing. With dissertations, my bias is to insist that the student come up with the initial idea, write it up etc.
 

Hope this helps.


Date: Sun, 14 Nov 1999 09:16:48 -0500

I personally never take first authorship on student papers -- I do not value first authorship (I teach history & systems and know it is not a reflection of scientific contribution). I would, however, not give the original student first authorship, might put the second student as first and the first as last (or, just following my tendency leave myself as last).
 


Date: Sun, 21 Nov 1999 06:19:59 -0600

I have just reviewed the series of comments on a interesting question of authorship credit. I plan to give my students an anonymous digest of this correspondence for discussion in their ethics class.
 

One point I did not see represented in the discussion is the idea of a written agreement on authorship at the time the supervisor agrees to take on the student.
 

I sign a written agreement with all my students (Honours BA, MA, and PhD) before starting a project that says, among other things, that if the project produces publishable results and the student completes the thesis AND writes at least a first draft of an article within 6 months of submitting the thesis, then she or he will be first author. After the agreed upon time period, if the student has not produced the draft, authorship is left to my discretion. I might write up the report and still give the student first authorship, depending on the balance of contributions, but at least the decision making process is clear and no debate is needed.
 

These agreements have helped considerably and the students seem to like them. They also cover things like frequency of meetings and relationship to my research team.

-CvB