Re: Reasons for freeing the primary research literature

From: Albert Henderson <chessNIC_at_COMPUSERVE.COM>
Date: Mon, 20 Aug 2001 16:46:44 -0400

on Fri, 17 Aug 2001 Stevan Harnad <> wrote:
> On Thu, 16 Aug 2001, Albert Henderson wrote:
> > Instead of scientific studies to support
> > the misnamed "self-archiving" argument, we are abused
> > with the rhetoric and nonsense such as attempts to
> > justify the phrase "virtually all" while citing a
> > source that provides the statistic "36.87%."
> I patiently repeat that the "virtually all" refers to the proportion of
> self-archived preprints in the Physics Archive that are submitted to
> refereed journals. The respective acceptance rates of those journals
> are a separate (and completely irrelevant!) matter.
        I simply quoted the very source that Stevan cited to support
        his myth-inspiring phrase. I also note that the thirty-percent
        range was consistent with other research on formal publication
        of informal papers

> The 36% referred to the number of authors that updated their reference
> at that time: this is another irrelevant statistic (for Albert's
> purposes), about which the author, Tim Brody, has already posted a
> response to this Forum.

        Now that this source is clearly involved in a
        propaganda campaign where conclusions are so often
        unrelated to the facts, who would take it seriously?

> > Support for "self-archiving" is made more foolish by
> > the fact that, as even its most ardent supports in this
> > forum have pointed out, authors are notoriously difficult
> > to regulate. Whatever is made public outside peer-reviewed
> > journals cannot be trusted as a general rule. Moreover,
> > no one can guarantee that charlatans will not insert
> > counterfeit claims of research to support their private
> > commercial interests.
> Albert predictably keeps speaking of self-archiving as if it were the
> self-archiving of unrefereed research, whereas this is all about the
> self-archiving of refereed (= peer-reviewed), published papers. The
> pre-refereeing preprints are merely a bonus, over and above the
> refereed postprints.

        The point is that no one can be expected to tell the
        difference. Moreover, Steven and others so often use
        preprint statistics (see above) to apply their
        arguments to the formal literature, I am certain
        of this.
        There are other observations along these lines. For
        example, Callaham et all commented, "Presentation of
        scientific studies at meetings is an important part of
        the dissemination of knowledge, but half of these studies
        appear only as abstracts and never undergo any other peer
        review. Whether the abbreviated peer review used to select
        abstracts for meetings actually identifies scientific merit
        is unknown, yet abstracts are cited as often as fully
        published papers."

        [Callaham, M.L., et al Positive - outcome bias and other
        limitations in the outcome of research abstracts submitted
        to a scientific meeting. J A M A 280:254-257 1998]

> I think it would be useful if Albert reviewed the logic of
> conditional probabilities: From the fact that many papers are first
> self-archived at their pre-refereeing preprint stage (in Physics)
> it does not follow that the later (refereed) stage (1) does not take
> place or (2) is not self-archived!

        The independent research indicates that most research
        offered as informal papers (preprints, conference
        papers, etc.) never sees the light of day as an article
        in a refereed journals. These are more empirical
        observations than a "conditional probability."

        Not long ago, Weber et al asked, "Why are the results of
        many studies never published? Our studies confirm prior
        reports that most unpublished research is never submitted
        to a journal for review. Only 20% of the unpublished studies
        originally submitted to the SAEM meeting were later submitted
        as a full manuscript to a journal. Moreover, investigators
        were easily dissuaded, submitting a manuscript, to fewer than
        2 journals before giving up.

        [Weber, Ellen J., et al. Unpublished research from a
        medical speciality meeting. JAMA 280,3: 257-259. 1998]

        In 1969 Belver Griffith reported that about 25% of the
        authors of printed papers delayed or decided against
        submission of a paper based on their convention presentation.
        Replication of the trial the in 1966 and 1967 eliminated the
        "free" distribution and "novelty" interest indicated that
        authors were still satisfied with Proceedings as a means of
        publication and that prepublication distribution of papers
        improved the exchange of information.

        [American Psychological Association. 1969. Reports of the
        Project on Scientific Information Exchange in Psychology, vol. 3.
        Washington DC, American Psychological Assn. 261 pp. NTIS PB 182962]

        Garvey and Griffith found "two thirds of the technical reports
        produced in 1962 had not achieved journal publication by 1965, and,
        apparently, the contents of the vast majority of these reports were
        never submitted for journal publication. Many authors of such
        reports indicated that 'no further dissemination of the information
        was necessary.' ... This raises some questions about the ultimate
        value of the information in these reports and its relevance to the
        established body off scientific knowledge."

        [W D Garvey COMMUNICATION THE ESSENCE OF SCIENCE Pergamon 1979. p. 136]


Best wishes,

Albert Henderson

Received on Wed Jan 03 2001 - 19:17:43 GMT

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:46:13 GMT