To oblige David's request, I found both the Wren study and the Antleman's
study to be excellent and defendable studies. Both of them defined Open
Access as some form of reprint found somewhere on the Internet other than
the publisher's site (author's own site, repository, etc.). In other
words, they were not interested in Open Access as primary publishing
model, but on enhanced availability afforded by authors republishing (or
redistributing) their own work online.
Both also reported a similar result, that is, there was a high degree of
association between high-prestige journals and frequency of author
reprints. In Wren's study, journals with high Impact Factors (New England
Journal of Medicine, Nature, Science, and Cell) were associated with a
higher degree of author republishing than lower-impact journals. He goes
further and to discusses possible causes of this difference and briefly
discusses a "trophy effect" -- the desire for researchers to display their
accomplishments-- which would explain why high impact publications are more
common online" (p.4). Antelman also suggests from her data that "the
greatest impact of open access is with the most-cited articles" (p.378).
In effect, there may be two complimentary processes taking place at the
same time: 1) A self-promotion effect (Wren's Trophy Effect, where authors
are more likely to promote their own high-impact articles) 2) The Mathew
Effect (where readers are more likely to cite high-impact articles)
If we take these two axioms as being true, then generalizations (like open
access publishing increases citations from 50%-250%) should not be made
without sufficient qualifications. It may be more reasonable to say that
"author republishing/redistribution may increase citation impact,
especially among highly prestigious journals and authors".
--Phil Davis
SOURCES
Merton, Robert K. "The Mathew Effect in Science", Science. Jan 5, 1968
159(3810):56-63
Can be found in JSTOR