Readers may want to see the commentaries on the Davis & Fromerth (2007) article at:
http://www.ecs.soton.ac.uk/~harnad/Hypermail/Amsci/subject.html#5221
In summary, no study has yet succeeded in eliminating the possibility of
some Self-selection Bias contributing to the Open Access (OA) citation
advantage, but no study has shown that all or most of the OA citation
advantage is dues to Self-selection Bias either. (Self-selection Bias,
SB, or Quality Bias, QB, would occur if authors selectively tended to
self-archive the articles that they felt were of higher quality. Eysenbach
(2006) asked authors whether this was the case, and they said no, but
of course that doesn't quite settle the matter either!
http://biology.plosjournals.org/perlserv?request=get-document&doi=10.1371/journal.pbio.0040157
).
In particular, Davis & Fromerth certainly have not settled the
matter. They found no Early Access advantage in the subset of journals
in mathematics that they analysed, and this could well be, for a
long-turn-around-time discipline like mathematics, compared to shorter
turnaround fields like physics and biology (though subfields differences
might over-ride this in some cases).
Davis & Fromerth also showed that in his sample of math journals,
articles deposited in Arxiv (hence OA) had 35% more citations than non-OA
articles, and that the proportion of OA to non-OA articles was bigger for
more highly cited articles. That certainly doesn't demonstrate that
the OA advantage is due to self-selection! It is just as likely that the
advantage of being OA is greater for higher quality articles! (Conversely,
for low quality articles, making them more accessible won't help!) (We
found the same correlation in our studies of a million and a half articles
across 12 years and 10 different disciplines):
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/
But because Arxiv is a central archive with a 15-year history of joint usage by the
field, the OA advantage is confounded with what Michael Kurtz has called the "Arxiv
advantage" (AA): Many users will prefer to consult the Arxiv version rather than the
publisher's version, even when they have subscription access to the publisher's
version, because they prefer a one-stop-shop. The Arxiv advantage probably explains
the lower number of downloads at the publisher's site for papers self-archived in
Arxiv.
Harnad, S. (2005) OA Impact Advantage = EA + (AA) + (QB) + QA + (CA) + UA.
http://eprints.ecs.soton.ac.uk/12085/
Stevan Harnad
On Tue, 27 Jun 2006, Phil Davis wrote:
> In our study of math articles deposited in the arXiv, we could not detect
> an Early View advantage. Mathematics articles have very long citation
> half-lives and don't get cited nearly as often as biomedical articles, so
> the effect may be there, but just not detectable in our dataset. There
> were stronger explanatory variables to explain the citation advantage. See:
>
> Does the arXiv lead to higher citations and reduced publisher downloads for
> mathematics articles?
> Philip M. Davis and Michael J. Fromerth
> Scientometrics (2007 forthcoming) http://arxiv.org/pdf/cs.DL/0603056
>
> An analysis of 2,765 articles published in four math journals from 1997 to
> 2005 indicated that articles deposited in the arXiv received 35% more
> citations on average than non-deposited articles (an advantage of about 1.1
> citations per article), and that this difference was most pronounced for
> highly-cited articles. The most plausible explanation is not Open Access or
> Early View, but Self-Selection, which has led to higher quality articles
> being deposited in the arXiv. Yet in spite of their citation advantage,
> arXiv-deposited articles received 23% fewer downloads from the publisher's
> website (about 10 fewer downloads per article) in all but the most recent
> two years after publication. The data suggest that arXiv and the
> publisher's website may be fulfilling distinct functional needs of the reader.
Received on Tue Jun 27 2006 - 18:08:39 BST