6

Normally, the credit of a scientific paper is equally distributed among authors. In the real world, it is almost impossible to write a paper with equal contributions of all authors. Usually, one author is the main one who is the idea maker or discoverer by subtle analysis of data.

In most cases, contributions of some authors are trivial, e.g., consultation on a specialized area, analysis of a part of data, etc. With this system, there are many famous professors who are co-authors for tens of hundreds of papers.

Why when judging the research records of a researchers, only the quantity and quality of papers are considered, and no one cares about how much s/he has contributed to those papers?

15
  • 3
    It's not perfect. Have you got a better idea? Sep 17, 2013 at 23:01
  • 1
    @NateEldredge not a practical one, but I met a promotion committee who needed a statement for contribution to each paper, as major journals like Nature recently add to each paper. However, it is just a note at the end of papers with no real considerations within academia. I also witnessed that some committees significantly credit the first author as the main author.
    – Googlebot
    Sep 17, 2013 at 23:11
  • This is not true where I teach. We are expected to quantify and describe our contribution if we are not the sole author of a paper.
    – Joe Hass
    Sep 17, 2013 at 23:21
  • @JoeHass describe to whom? When someone posts his resume on personal webpage or submits for a job application, everyone just look at the list of publications and do not bother with his actual contributions.
    – Googlebot
    Sep 17, 2013 at 23:28
  • 1
    @JoeHass Just a personal example of interviewers asking about ones contribution to a paper: I didn't just get asked "How much did you do?" I was asked to "Please describe my contribution in some detail". If I just said that I did everything, that would have been an obvious lie, as I couldn't talk about details of about half a project. My part I could describe perfectly and explain how it complemented the results of the other part (that's the only thing I knew). They won't ask this for every one of your papers, but could ask about any of them.
    – penelope
    Sep 18, 2013 at 7:33

3 Answers 3

5

I agree that using simple publication or citation counts is prone to a range of problems. One important class of such biases relates to differential levels of involvement of an author.

In terms of some basic corrections for this, there are ways that individual papers are evaluated to assess relative contribution.

  • In some fields being first author (and to a lesser extent, subsequent positions) suggests a greater level of involvement in a given paper.
  • In a few cases, proportion of contribution is explicitly stated.
  • In some contexts, some form of proportional analysis is performed such that if you are one author among two, then the paper is weighted more than one author among ten (I've seen this in some university workload models).

However, generally, when evaluating the publication achievements of a researchers, a more holistic approach can be taken.

  • An overall program of research will be evaluated in order to assess a sustained contribution.
  • To some extent, over time researchers would be expected to play a range of roles from lead to secondary contributor. Therefore, often an overall sum of papers may balance out such varying levels of contributions. In particular, if you have a reasonable balance of first author papers and the number of authors per paper is similar to other researchers in your field, this may reinforce such a perspective.
10

My opinion is that, contrary to what you say about "one main contributor", in most cases, it is absolutely impossible to judge "the level of contribution" from any objective standpoint. Let's consider a (not hypothetical, just missing real names!) example.

A and B spent about a year thinking of a problem and devising a scheme for solution they could not make work.

A discussed the question with C.

C, who wasn't much interested in trying it yourself, passed it to his collaborator on a different project D.

D found an approach that gives fairly good result but not quite what was wanted and told it to C.

C passed it to A and B and during the two week visit of D arranged that A,C,D have a few discussions about it that resulted in some extra ideas but not a full solution yet.

Meanwhile B tried to combine what he knew himself and what he was told by C and obtained an even slightly better result than D (though still short of the exact statement wanted) but under more restrictive assumptions. He sent it to D.

D, upon reading B's draft, realized that the initial scheme of A and B could be made to work after all (what he was missing was in that note from B and what B was missing was a part of D's "general knowledge"), finished it off, and sent the solution to A,B,C, who have read and verified it.

Now, I suggest you try to tell the "level of contribution" of each person keeping in mind that

1) If not for C, D would most likely not hear of the problem at all and it is doubtful that he would come into direct contact with A,B.

2) It is possible that A and B would make their approach work eventually without D.

3) What was a "general knowledge" for D and allowed him to finish the problem off, would hardly come into the mind of A,B,C at all.

4) Without B's draft, D would, most likely, stop at the "partial result" he obtained first.

5) Both A and C participated in the discussion during D's visit and, while everybody remembers all the ideas that surfaced, nobody remembers (or cares much about) who said what.

6) A,B,C,D all argue that the rest 3 could surely do the problem without him, just in longer time, so the current idea of "author credit" C and D have is to publish the whole thing under the name AB CoD.

0

I have to disagree that all authors are credited equally. Since there are several ways in which authorships are counted, the position of your name will provide different signals. In my realm, first authors are simply assumed to have done the most, such papers count higher. Other authors are considered to have done less in falling order, unless stated otherwise. In some fields the last author is the "important" one. In some fields authorships come through contract, e.g. CERN consortia authorships (see link in the reply on coauthorships, the answer there may also be of help). Hence authorships are treated differently.

Ideally everyone should follow the Vancouver Protocol) in which it is clear that some contributions simply should not warrant authorship but rather a mention in the acknowledgement.

So in the end when judging the authorships of papers, each field applies their own "standard".

1
  • 3
    follow the Vancouver Protocol: "Editors may ask authors to describe what each contributed; this information may be published." After reading this, I lost quite a noticeable part of my belief in the assumption that those guys have clear idea of what they are talking about and are in touch with current realities and customary practices of collaboration.
    – fedja
    Sep 18, 2013 at 15:46

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .