Re: Journal rejection and acceptance rates

[Joseph Esposito <espositoj@gmail.com>]

These questions are all reasonable, but it is difficult to answer
them, as one literally has to begin at the beginning.

I am going to take a partial stab at this, but before I do, to
some extent it has to be said that the questions derive from the
red herring of the quantitative measure of rejection rate.  It's
a bad number, like all metrics that are supposed to say something
about quality (e.g., impact factor), part of the unfortunate
attempt to make things somehow sound more scientific because they
are wrapped in numbers.  The numbers (which are metaphors for
quality and sensibility) become the currency of the trade, with
publishers boasting of their high figures and librarians making
purchasing decisions based on them.  But at some point all
metaphors break down. My love is like a red rose, except when she
isn't.

I have noted on this list before that the purpose of editorial
selection is to suppress production.  There is much more written
in any field than can be read and absorbed by anyone.  Editorial
selection is partly a filter; it helps determine what is worth
reading.  The role of the brand (Lancet, JACS, NEJM) is to say
that you the reader have come to appreciate the choices that a
particular publication has made.  Those choices save readers
time.

But editorial work is also about sensibility.  Authors may not
like this.  A good article may not be published because a better
one has taken its place.  Or the editor is interested in having
the journal stake out a position in one area over another.
People working in a particular field know what these
sensibilities are.  They say that a particular article seems
better suited for one journal than another, even if both journals
are in the same field.

The medium (print vs. electronic) has nothing to do with it.
The fact that you can publish more in digital form doesn't mean
people have more time to read things.  It is a great disservice
to readers for a journal to advertise that it is the largest in
the world.  Of course, for some journals, the service is designed
for the benefit of authors, not readers.  This is the basis of
open access publishing.  There is of course nothing wrong with
doing things for the benefit of authors. It's simply a different
game than the value created for readers, which is what editorial
selection is geared for.

I have written about this before in the Journal of Electronic
Publishing:  http://bit.ly/4JKRMt.  Of course, someone may
determine that author-friendly publication (I think of it as
"supply-side" publishing) is a superior form, and it may very
well be.  Fortunately, we don't have to choose.

Joe Esposito

On Mon, Jan 31, 2011 at 3:28 PM, Ken Masters
<kmasters@ithealthed.com> wrote:

> Hi All
>
> I'd like to ask several questions about a topic that rears its
> head from time to time on this group, but doesn't always appear
> to be dealt with in detail: journal rejection and acceptance
> rates.
>
> 1.Why are rejection rates advertised as some measure of quality
> - or at least they seem to be, when advertised and discussed on
> discussion groups? (I can understand the logic of the argument
> to a limited extent, but it means that a journal with a 100%
> rejection rate is the best of its type in the world.)
>
> 2.Similarly, given that a paper can be rejected for many
> different reasons and that rejection rates are affected by a wide
> range of factors, mostly due to (variable) editorial policies
> (physical space, plagiarism, inappropriate subject (too specific,
> not directly related, etc), withdrawal by the authors, number of
> revisions required, number of reviewers' rejections required for
> a rejection of the paper, whether or not reviewers are anonymous,
> number of invited papers, etc), why does quality seem to be the
> only thing implied when people discuss rejection rates?
>
> 3.When journals publish (actually, advertise) their rejection
> rates, why do they not routinely break down their rejection by
> cause?
>
> 4.Given that journals with a high rejection rates tend to also
> receive a large number of low-quality papers, there is a point at
> which the rejection rate become self-inflating and even less
> related to quality. What is that point?
>
> 5.In my experience as a reviewer, I have seen many good papers
> rejected because of lack of physical space, and some journals
> make this point on their websites. If space is a problem with
> paper-based journals, and not a problem with online journals,
> then, all other things being equal, doesn't it stand to reason
> that paper-based journals will almost always have a higher
> rejection rate than online journals? Again, if there were a
> breakdown of reasons for rejection, this would help to clarify
> the discussion.
>
> 6.If a rejection rate is to be some measure of quality, then what
> is the optimum rejection rate?
>
> 7.Given that, when comparing percentages, the statistics for
> _anything_ don't mean significance unless you know the raw
> figures, does it even vaguely make sense to say that one journal
> has rejection rate of X% and another has a rejection rate of Y%
> unless you also cite the raw data?
>
> 8.Are rejection rates _ever_ externally audited and verified, the
> way that the publishers' financial books are audited and
> verified? If not, shouldn't they be, given the tendency to
> equate quality with rejection rate?
>
> 9.Is there any universal method of calculating acceptance rates?
> e.g. if a journal has received 100 papers, 20 of which have been
> rejected, 30 accepted, and 50 are still under review. Is this an
> acceptance rate of 30/50 =3D 60% or an acceptance rate of 30/100 =
> 30%? How are withdrawals factored into this?
>
> And this doesn't cover the issue of differences in rejection
> rates across disciplines.
>
> Disclaimer: I'm EiC of an open-access online journal, and one of
> the stated aims is to _publish_ rather than reject. My reviewers
> are encouraged to advise authors on how they can improve their
> papers in order to get it to publishing level.
>
> Regards
>
> Ken
>
> Dr. Ken Masters
> Asst. Professor: Medical Informatics
> Medical Education Unit
> College of Medicine & Health Sciences
> Sultan Qaboos University
> Sultanate of Oman
> E-i-C: The Internet Journal of Medical Education