Re: More on Open Access citations

[Stevan Harnad <harnad@ecs.soton.ac.uk>]

The M & S article needs to be read more clearly. This is not 
about open access, it is about online access. Journals have gone 
online at various intervals in the past 15 years. Using a variant 
of Evans and Reimer's (rather flawed) methodology, M & S tested 
whether the increased ease of access -- online vs on paper -- 
made any difference to citations, and it didn't.

This has nothing whatsoever to do with Open Access. OA is about 
providing access to those who don't have subscription access; 
this paper is about providing online access to those who had 
subscription access before, but on paper, and now have online 
access. Moreover, it is in a field (economics) where there is 
wide posting of OA preprints (not taken into account at all).

Conclusion, in increasing *ease* of access for subscribers (by 
making subscribed content accessible to subscribers online 
instead of just on paper) does not increase citations (in a field 
where preprints are widely available OA).

This says nothing about OA, where the free online access only 
makes a difference to users who do *not* have institutional 
subscription access: *That*'s where the OA citation advantage 
comes from.

The authors (for some bizarre reason -- probably because their 
economic modeling work is done mostly from the publisher's point 
of view) draw the conclusion that it is not in the author's 
economic interest to pay extra to publish in an OA journal (where 
did OA journals come into this?) because it does not buy him more 
citations.

The evidence for the OA advantage comes from comparing articles 
in the
same (*subscription*) journal, same year, that have and have not 
been
made OA by their authors.

M & S (for unknown reasons) cite and repeat Phil Davis's 
non-replication
of the OA advantage in his small sample of biology journals. They 
don't
mention recent studies, including the one showing that the 
mandated OA
advantage is just as big as the self-selected OA advantage. So 
they
simply repeat the hypothesis that the OA advantage is just a
self-selection advantage because of the tendency of better
authors/papers to be made OA...

Chrs, Stevan

On Wed, 9 Feb 2011, David Prosser wrote:

> The absolutely fascinating thing about the paper by Mark McCabe
> and Chris Snyder is that it appears to show that there is, in
> general, no citation advantage accruing to online content.  Not
> just to online, open access content, but to any online content.
> Phil makes this clear in the title of his blog post, and it is
> only here that the subject is narrowed somewhat to only open
> access citations.
>
> If McCabe and Snyder are right then the widening of access
> through big deals, third-party intermediates, and open access has
> made no difference to citation rates.  Are access and citations
> so disconnected that an increase in one has no effect on the
> other?   Is it really the case that none of the business models
> for online access have increased citations?  Intriguing.
>
> (There is a JSTOR twist in the results; read the paper for
> details.)
>
> David
>
> On 9 Feb 2011, at 00:20, Joseph Esposito wrote:
>> The question of whether or not Open Access leads to more
>> citations has come up again, this time in listmember Phil Davis's
>> blog post:
>>
>> http://bit.ly/gHpQF0
>>
>> Since OA is as much a theological debate as it is a property of
>> one kind of publishing, it is hard to have reasoned discussion on
>> the topic.  Davis is a data hog of the first order and has to be
>> taken seriously.  The notion that Davis is somehow "anti-Open
>> Access" is nonesense.
>>
>> Davis and I don"t agree about this matter on what I would call
>> temperamental grounds. I am myself uneasy with what I see as
>> excessive quantification, asking numbers to do what only human
>> judgment can.  I wrote about this some time ago:
>> http://bit.ly/8z1yMM.  Citation analysis gets you to the front
>> door, but it doesn"t let you inside.
>>
>> Most people who disagree with Davis, however, also disagree with
>> me. So this is not a binary argument.  I remain puzzled that for
>> all the benefits of OA publishing, advocates continue to cling to
>> two arguments that are simply not true, that it lowers costs (to
>> whom?) and that it increases scientific impact (no evidence).
>> Why not discuss OA for what it does remarkably well?  It provides
>> a vehicle for a much broader display of research materials, lends
>> itself to emerging data-mining techniques, enables individual
>> authors to take greater control of the publication of their own
>> work, and provides a public venue for work that is highly
>> specialized and not likely to find a market by traditional means.
>> By these standards, OA is a home run.
>>
>> Joe Esposito
>>
>