Page 68 - Becoming Metric Wise

58 Becoming Metric-Wise

observed effect? In an attempt to answer this question Moed (2007) com-
pared the citation impact of articles deposited in the ArXiv and subse-
quently published in a scientific journal, with articles published in the
same journal and not deposited in the ArXiv. To discriminate between an
OA effect and an early view effect he performed a longitudinal study over
a period of 7 years. Author self-citations were excluded. His study pro-
vided evidence for a strong early view and quality effect. Correcting for
these no sign of an OA effect was found. Note that Moed investigated the
Green OA case. The next study investigates Gold OA.

3.2.8 Davis’ Randomized Controlled Trial on Gold OA
If OA articles receive more citations, then the question is why? Davis
et al. (2008) made an arrangement with a group of publishers to ran-
domly pick some articles and make them freely accessible while other
ones in the same issue were not. In total 36 journals in the sciences, social
sciences, and humanities were involved. Neither authors, editors nor pub-
lishers were involved in this random assignment (otherwise it would not
be random). They found that articles in OA received significantly more
downloads and reached a broader audience (in the sense of more unique
IP addresses) within the first year than subscription-access articles in these
same journals. However, they found that OA articles were cited no more
frequently, nor earlier, than subscription-access articles within a 3-year
period. This finding contradicted many others. Yet, these others did not
include the randomness feature. Davis et al. explained their result by
social stratification. The best scientific authors are concentrated in a rela-
tively small number of research facilities with excellent access to the sci-
entific literature. They concluded that the real beneficiaries of OA are
those who do not belong to this scientific elite: those who publish (and
hence cite) less or not at all (students, practitioners, retired faculty).
In order to correct for selection bias, a new study by Gargouri et al.
(2010) (including Harnad) compared self-selective self-archiving with
mandatory self-archiving in four research institutes. They argued that,
although the first type may be subject to a quality bias, the second can be
assumed to occur regardless of the quality of the papers. They found that
the OA advantage proved just as high for both, and concluded that it is
real, independent and causal.
Moed (2012) mentioned that there is a bias when using citations in
important, international journals covered by databases such as WoS or