[AISWorld] A suggested resolution for multiple submissions

Wed Dec 14 15:53:17 EST 2011

Dear colleagues,

I have followed this discussion with interest, especially because I have some particular perspectives on "self-plagiarism". However, I will address that aspect directly in a separate post. Here I want to suggest a resolution for the problem of multiple submissions.

I want to first commend Robert Davison and the team on the AIS Research Ethics Committee for their efforts and accomplishments in the challenging task of dealing with academic plagiarism. Dealing with cheating with students is hard enough; when academics do it, the complexities and gravity is greatly multiplied. I am a recent "victim" of academic plagiarism (though I find "victim" an awkward word to describe it, as my first reaction was feeling honoured that my work was considered worthy to be plagiarized), and I realize that these issues are serious and extremely delicate. I thank and commend them for their labour on behalf of the IS community.

It seems clear from the discussion that multiple submission is very sticky to define, and indeed, is inconsistently defined, depending on the publication target, the editor's opinions, and the reviewers' opinions. This greatly complicates authors' responsibilities, as what might be acceptable to one target might be considered unethical by another very similar target. It also becomes quite delicate for those who after the fact have to make a decision on whether unethical multiple submission was attempted or not. Actually, I don't like the idea of a one-size-fit-all policy coming down from AIS that would try to authoritatively define what is "multiple submission"; I believe different conference, journals, books, etc. can legitimately make different definitions depending on their unique goals. I don't think an authoritative definition would really solve the problem.

I suggest a simple way to get around these murky issues: rather than leaving it to authors to make sure they avoid multiple submission, the editors should take responsibility by explicitly clarifying what they consider "multiple submission." Whenever a conference, journal, workshop or any other publication outlet solicits submissions, they should do two things concerning multiple submission:
1) Explicitly state what they consider acceptable and what they consider unacceptable for their publications. This is very similar to Stephen Alter's point #3, which I will not explicate further here.
2) Require authors to explicitly list the full citation of all publications that share more than 20% of the same content, and explicitly explain what novel contribution the present submission offers beyond previously published content. (The 20% threshold is arbitrary and might be somewhat low, but I believe it should cover anything that might be of reasonable concern.)

With this simple solution, the editors can decide what constitutes multiple submission before they send the submission on for review. Thus, it is no longer the responsibility of authors to determine grey areas, but rather the editors'. This solution offers clarity for three kinds of people or situations:

1) For sincere authors who are generally aware of the problem of multiple submission and who are committed to not violate any ethical principles, this solution lets them be upfront about their submission, and leave the decision of acceptability wholly in the hands of the editors.

2) For authors who sincerely do not really understand the problems of multiple submission and who would otherwise not consciously act unethically, this solution serves as a valuable educational experience with no risk of embarrassment.

3) For unethical authors who are trying to hide multiple submission, this solution leaves them in a quandary. They could go ahead and make an unethical multiple submission and hide the fact. However, if discovered, they would be unable to claim ignorance or misinterpretation as an excuse for their deceit. Alternatively, this solution would dissuade them from submitting to outlets that make such requirements.

In fact, I personally always practice this principle: In my submission cover letters, I always cite prior related publications and explain their relationship and difference with my present submission. This has helped me in at least one case where I submitted an article for consideration at a leading scholarly journal based on a student thesis where the student had already published the thesis as a book. After a few months, I received a response from the editor-in-chief that an anonymous reviewer had identified the book as a prior publication. I referred the EiC to my cover letter where I had explicitly mentioned this fact. Eventually the EiC decided that prior publication in a book was inappropriate for his journal that sought original research; however, he made it clear that we were welcome to submit other articles in the future. We eventually published the article in another journal with similar quality standards (again explicitly citing the book in the cover letter), and 
no issue was ever raised. (A side note on copyright: the book publisher explicitly permitted republication of the content in scholarly journals, so the issue was only one of originality, not of copyright conflicts.) My point with this anecdote is that what constitutes "multiple submission" is not black-and-white; two editors of similar journals can hold different standards. As long as authors are upfront so that no deceit is suspected, then any misunderstanding can be clarified relatively easily.

In brief, I recommend that editors explicitly define what they consider "multiple submission" for their specific outlet, and that they ask authors to list all prior provenance of their submissions; I recommend that authors always provide prior provenance of their submissions whether the editors request it or not. This might not solve the problem once and for all, but it is very easy to implement (at least before an alternative better long-term solution might be implemented), and I think it would greatly alleviate the problem immediately. That is, it could rapidly deter unethical multiple submissions, and make it easier to prove the unethical nature when violators are caught.

Regards,

Chitu Okoli
Associate Professor in Management Information Systems
John Molson School of Business
Concordia University, Montréal

http://chitu.okoli.org/pro

Message: 4
Date: Mon, 12 Dec 2011 16:51:31 +1100
From: Roger Clarke<Roger.Clarke at xamax.com.au>
To:aisworld at lists.aisnet.org
Subject: Re: [AISWorld] Follow-up to AIS Insider: Plagiarism: Your
	Input Sought
Message-ID:<p06240802cb0b08d3faa0@[192.168.1.100]>
Content-Type: text/plain; charset="us-ascii" ; format="flowed"

Below are responses and references to several aspects of the discussion.

(Declaration:  I was a member of the 3-person Task Force that
prepared the current version of the Code of Research Conduct;  but I
haven't been involved in the current process).

1.  Guidelines for Professional Practice

One posting suggested that we need "guidelines for professional
practice in the submission of academic articles to journals and
conferences".

AIS has them (and has had since 2003, revised 2009).
They're in the 'About AIS' drop-down menu, 7th bullet, at:
http://home.aisnet.org/displaycommon.cfm?an=1&subarticlenbr=15

Naturally the current version isn't the last word;  and the recent
problems and the current discussion are very likely to result in
refinements.

2.  Why Is Parallel Submission a Bad Thing?

Parallel submission results in duplication of the work done by
editors and reviewers.  That's a serious problem, because quality
assurance is utterly dependent on pro bono work by senior members,
and anything that wastes their resources and reduces their commitment
results in lower quality.

Note that at least Rick Watson and myself have argued on multiple
occasions that technological change during the last 15 years has
prepared the ground for us to re-engineer the submission-and-review
process.  See:
http://www.rogerclarke.com/EC/ePublAc.html#RevP

3.  Administrative Abuse of Plagiarism Accusations

The case that Ned Kock drew attention to is galling.  An earlier case
study together with a set of evaluation criteria for text-books are
here:
http://www.rogerclarke.com/SOS/Plag0602.html#CS

4.  A Better Term Than Self-Plagiarism

At 13:09 +0100 11/12/11, Key Pousttchi wrote:

> The term "plagiarism" is a very sharp sword in public discussions ...
>   ... I would urge that we think about a different
> term for the second, e.g., something like "double-selling".
The Code intentionally avoids using the term 'self-plagiarism' -
except for mentioning that the expression is sometimes used for
"unimproved re-publication of one's own work".

A more precise definition would be better.

The following was proposed at:
http://www.rogerclarke.com/SOS/SCSP-09.html#RTFToC8
"'the re-presentation of the documented words or ideas of oneself,
without appropriate attribution"

The criteria of 'naughtiness' are:
-   re-presentation;  boolean-&
-   non-disclosure of prior presentation

And value-add is not enough to escape.  Even where an 'improved'
re-presentation is offered, the previous publication must still be
cited.

Referees (if they aren't blind), and in any case editors, can then
evaluate the paper on its merits, taking into account the fact that
it contains some 're-presentation'.

5.  Blind Review

At 8:06 +0100 10/12/11, Ilia Bider wrote:

>   ...  When a work is based on own research, published in accepted
> sources or on Internet, it is often needed to repeat parts of the
> previous works in a new paper, otherwise it becomes unreadable
> Isn't it time blind submissions were abandon?
The problem of 'reviewer blindness' is analysed here:
http://www.rogerclarke.com/SOS/SCSP-09.html#RTFToC15

The blind review approach derives from the days when it was thought
that objectivity was attainable.

Both authors and reviewers have responsibilities.  Quality is far
better served through transparency, not obscurity.  Double-open is a
much more mature way to carry on business.

(Both authors and reviewers retain the capacity to have confidential
discussions with the editor when necessary;  but the default needs to
be that communications are open among editor(s), author(s) and
reviewers).

6.  How to Define a 'Prior Publication'

Some kinds of prior exposure disqualify a paper from being submitted
to a refereed venue, whereas others kinds of prior exposure do not.

This is addressed in the Code at para. 4:
http://home.aisnet.org/displaycommon.cfm?an=1&subarticlenbr=15#Category_Two

For example, a 'Departmental Working Paper' or an exposure draft
(whether on the author's own web-site or the site of a professional
body such as Sprouts) do not represent a 'Prior Publication', whereas
a refereed conference is, and so is a journal in another discipline,
and so is a book-chapter.

As Claudia Loebbecke points out, there are interpretation problems
with 'research-in-progress', and with 'pre-events' to major
conferences.  The Code makes allowance for the existence of a grey
zone, and requires people who are making judgements to take full
account of the context.

The proposal uses the following test:  "has been published elsewhere
in a workshop, seminar or other venue ***not protected by
copyright***".

Nice try;  but copyright-ownership is simply irrelevant.  The quality
of a venue is determined by its editors, reviewers and submitters;
not by its publisher, and certainly not by its publisher's copyright
policies.

As per Krassie Petrova"<krassie.petrova at aut.ac.nz>:

> [Aren't] 'copyright' and plagiarism  ... different issues?
> In my view misrepresenting / appropriating others' work - whether
> copyright protected or not, is not acceptable as it is unethical
> with respect to academic ethical standards
A more appropriate way to express the distinction would be as follows:

'Ethical Guideline::
(1)  A paper may be submitted to a refereed venue even if it has been
previously published in a relatively informal venue such as a
workshop, a seminar, a departmental working paper, a personal
web-site, or a professional association web-site such as Sprouts.
(2)  A paper may be submitted to a refereed venue even if the content
has previously been the subject of a preliminary research-in-progress
report, whether that report was at a relatively formal or informal
venue.'

(Note that the expression above leaves conferences and journals free
to say that they will decline to consider some or all of those
categories of papers, or to set hurdles, such as asking reviewers to
ensure that the content is significant enough to warrant
re-presentation.  The role of the AIS Code is to provide ethical
guidance;  whereas the role of conferences and journals is to provide
formal venues for publications in accordance with their own
philosophies).

The following papers address these issues in depth:

Clarke R. (2006)  'Plagiarism by Academics: More Complex Than It
Seems'  J. Assoc. Infor. Syst. 7, 2 (February 2006), at
http://jais.isworld.org/articles/default.asp?vol=7&art=5, PrePrint at
http://www.rogerclarke.com/SOS/Plag0602.html

Clarke R. (2009)  'Self-Plagiarism and Self-Citation:  A Practical
Guide Based on Underlying Principles'  Commun. AIS 25 Article 19
(July 2009) 155-164, Special Section on Journal Self-Citation,
PrePrint athttp://www.rogerclarke.com/SOS/SCSP-09.html

Clarke R.&  Kingsley D.A. (2007)  'ePublishing's Impacts on Journals
and Journal Articles'  Journal of Internet Commerce 7,1 (March 2008)
120-151, PrePrint athttp://www.rogerclarke.com/EC/ePublAc.html

-- Roger Clarke http://www.rogerclarke.com/

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.aisnet.org/pipermail/aisworld_lists.aisnet.org/attachments/20111214/08bc5f69/attachment.html>