Re: On Proportion and Strategy: OA, non-OA, Gold-OA, Paid-OA

From: Stevan Harnad <harnad_at_ecs.soton.ac.uk>
Date: Mon, 15 Jun 2009 22:19:11 -0400

            SH: The fact that the vast majority of Gold
            OA journals are not

            paid-publication journals is not relevant if
            we are concerned about

            providing OA to the articles in the top
            journals.


      I simply did not know that OA aimed at articles only in
      the top journals. Tell this to our friends in India,
      South-Africa and Brazil, and you will see their reaction.


This completely misses the point of my posting, which was about the
often quoted (and correct, but equivocal) fact that most OA journals
do not charge a publication fee: True. But most of the top OA
journals do charge a publication fee (and most of the top journals
are not OA journals).

      OA is not only for the scientific élite... It might be
      time to separate quality science from élite science. 


The point has nothing to do with "eliteness." By the top journals I
meant the top quality journals. And quality is determined by
peer-review standards. Get peers in each field to rank the journals
in their field by quality (which does not necessarily mean impact
factor). Then see which proportion of the top 10% are OA compared to
the proportion of the remaining 90%. Then check which proportion of
the OA journals that are in that top 10% do not charge a publication
fee, compared to the proportion in the remaining 90%.

      And if OA were only for élite science, what would be the
      OA advantage? Élite science tends to be located in élite
      schools with reasonably well-stocked libraries. In such
      schools, the OA advantage becomes far less visible, as
      apparently demonstrated in some areas of cosmology, etc.


I couldn't follow all of that. 

But if the question is whether the OA advantage (higher downloads,
more citations) is evenly distributed across all articles, or across
all quality-ranges, the answer is decidedly not. 

Perhaps it is a 2nd-order effect of the Pareto/Seglen rule (that the
top 10-20% of articles received 80-90% of all citations) that the OA
advantage is mostly to the top 10-20% of articles. See 

      Hajjem, C., Harnad, S. and Gingras, Y. (2005) Ten-Year
      Cross-Disciplinary Comparison of the Growth of Open
      Access and How it Increases Research Citation Impact.
      IEEE Data Engineering Bulletin 28(4) pp. 39-47.
      http://eprints.ecs.soton.ac.uk/11688/ 


      and 


      Gargouri Y. & Harnad  S. (in prep.)
      http://www.crsc.uqam.ca/yassine/SelfArchiving/LogisticRegression.htm


Stevan Harnad

      -----Original Message-----
      From: American Scientist Open Access Forum on behalf of
      Stevan Harnad
      Sent: Mon 6/15/2009 3:02 PM
      To:
      AMERICAN-SCIENTIST-OPEN-ACCESS-FORUM_at_LISTSERVER.SIGMAXI.ORG
      Subject:      Re: On Proportion and Strategy: OA, non-OA,
      Gold-OA, Paid-OA

      On 15-Jun-09, at 1:12 PM, David E. Wojick wrote:

            Steve, for us non-experts in OA (this is not
            an OA listserv) can you

            explain briefly what Gold and Green OA are in
            these proportions?

            Especially Green OA in reference to
            proportions 1 & 7. They seem to

            be two different measurements. The vast
            majority of journals are GOA

            but the vast majority of articles are not.


            I don't see how your conclusions follow from
            these simple

            proportions, not without additional premises.
            Perhaps you can

            explain that.


            David


      David, with pleasure (and my apologies for assuming
      transparency). The
      proportions are,
      I think, very important not just for OA reasons, but for
      bibliometric
      reasons too.
      Please see the further explanations below. -- Stevan

                  As I do not have exact figures on
                  most of the 9 proportions I

                  highlight below, I am expressing
                  them only in terms of "vast

                  majority"

                  (75% or higher) vs. "minority"
                  (25% or lower) -- rough figures
                  that

                  we

                  can be confident are
                  approximately valid. They turn
                  out to have at

                  least one rather important
                  implication.


                  1. The vast majority of current
                  (peer-reviewed) journal articles
                  are

                  not Open Access (OA) (i.e., they
                  are neither self-archived as
                  Green

                  OA

                  nor published in a Gold OA
                  journal).


      A peer-reviewed journal article is Green OA if it has
      been made OA by
      its author,
      http://www.nature.com/nature/focus/accessdebate/21.html
      by depositing it in an Open Access Repository (preferably
      his own
      institution's OAI-compliant Institutional Repository)
      http://roar.eprints.org/
      from which anyone can access it for free on the web.

      A peer-reviewed journal article is Gold OA if it has been
      published in
      a Gold OA journal
      http://www.doaj.org/
      from which anyone can access it for free on the web.

      There are at least 25,000 peer-reviewed journals, across
      all fields
      worldwide.
      http://www.ulrichsweb.com/ulrichsweb/

                  2. The vast majority of journals
                  are Green OA.


      Of the 10,000+ journals whose OA policies are indexed in
      SHERPA/Romeo,
      over 90% endorse immediate deposit and immediate OA by
      the author
      63% for the author's peer-reviewed final draft (the
      postprint), and a
      further
      32% for the pre-refereeing preprint.
      http://romeo.eprints.org/stats.php

                  3. The vast majority of journals
                  are not Gold OA.


      Currently 4221 journals are Gold OA according to DOAJ

      (Note that the c. 10,000 journals in Romeo do not include
      most of the
      Gold OA journals, although these would all be classed as
      Green, and
      all Gold OA journals also endorse Green OA
      self-archiving. Romeo
      does, however, index just about all of the top journals.)

                  4. The vast majority of citations
                  are to the top minority of
                  articles

                  (the Pareto/Seglen 90/10 rule).

                  http://openaccess.eprints.org/index.php?/archives/474-guid.html


                  5. The vast majority of journals
                  (or journal articles) are not
                  among

                  the top minority of journals (or
                  journal articles).


                  6. The vast majority of the top
                  journals are not Gold OA.


                  7. The vast majority of the top
                  journals are Green OA.


                  8. The vast majority of Gold OA
                  journals are not paid-publication

                  journals.

http://www.earlham.edu/~peters/fos/2009/06/careful-confirmation-that-70-of-o
                  a.html


                  9. The vast majority of the top
                  Gold OA journals are paid-

                  publication journals.


                  I think two strong conclusions
                  follow from this:


                  The fact that the vast majority
                  of Gold OA journals are not

                  paid-publication journals is not
                  relevant if we are concerned
                  about

                  providing OA to the articles in
                  the top journals.


                  Green OA is the vastly
                  underutilized means of providing
                  OA.


                  The implication is that it is far
                  more productive (of OA) for

                  universities and funders to
                  mandate Green OA than to fund
                  Gold OA.


      There are somewhere around 10,000 universities and
      research institutions
      worldwide. So far, 51 of them -- plus 36 research funders
      -- have
      mandated
      (i.e. required) their peer-reviewed research output to be
      made Green OA
      by depositing it in an OA repository.
      http://www.eprints.org/openaccess/policysignup/

                  Stevan Harnad
Received on Tue Jun 16 2009 - 04:28:31 BST

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:49:48 GMT