Re: Citation analysis of author-choice OA journals

[Phil Davis <pmd8@cornell.edu>]

Ari, The method for calculating the "cost-benefit for each 
additional citation" is provided on page 7:

"Considering that authors are required to pay the publisher for 
the ability to make their article freely available upon 
publication, we calculated the estimated cost per citation from 
our model (Table 3).  To do this, we multiply the open access 
citation advantage for each journal (a multiplicative effect) by 
the impact factor of the journal to estimate the citation gain 
within the first two years after publication.  This citation gain 
is then divided by the open access fees levied by the publisher 
for this service."

For example, the multiplicative effect estimated for 
author-selected open access publishing in the journal 
Bioinformatics is 1.19, (which means a 19% advantage).  We 
multiply this by the impact factor of the journal (1.19 x 5.039) 
= 0.96 or rounded up to about 1 citation.  If the price paid by 
authors (or their grants) is $2,800 for non-member institutions 
or $1,500 for member institutions, we simply take the cost and 
divide it by the benefit (0.96 citations) to arrive at $2,925 for 
non-members or $1,567 for members.

Remember that these are articles that authors select for 
author-choice open access publication may be qualitatively 
different than subscription-track articles, so part the effect we 
are measuring may not be access.  In a randomized controlled 
trial of OA publishing, where the researchers (not the authors) 
select the publication track, we were unable to measure an effect 
of access on citations, see:

Open access publishing, article downloads, and citations: 
randomised controlled trial 
http://www.bmj.com/cgi/content/full/337/jul31_1/a568

Phil Davis


Ari Belenkiy wrote:
> Phil:
>
> Following your article: can you please explain what 
> "cost-benefit for each additional citation" is and how do you 
> measure it?