Re: The Cost per Article Reading of Open Access Articles

From: Stevan Harnad <>
Date: Tue, 20 Jan 2004 13:35:36 +0000 (GMT)

On Tue, 20 Jan 2004, Jonas Holmstr[ISO-8859-1] öm wrote:

> I assume that you think that the author-cost-per-reading is a paradox
> because it seems to relate a reader measure to the author.

Yes. Is it not more straightforward and comprehensible to say:

(1) OA is preferable to TA for the author-institution because it
maximises the impact of the author-institution's own research output
(and the rewards of impact).

(2) OA is preferable to TA for the user-institution because it
maximises the user-institution's access to the research output of all
other institutions.

(3) OA is more cost-effective than TA to institutions because: (a)
it costs less money and (b) it provides far more impact and access for
that less money.

But calculating the cost-per-download (to the user-institution) in OA
(when the cost is nothing) and comparing it to the cost-per-download
(to the user-institution) in TA (when the cost is real and substantial)
is (I think) needlessly straining to compare apples with oranges.

The cost-effectiveness of OA is best reckoned, and understood, as the
increased (indeed maximised) impact and access it provides. The monetary
measure of this is not download dollars, but the research and other
revenue that is generated by an institution's research impact (and,
more indirectly, the revenue from the enhanced research productivity of
researchers better informed about relevant research because they have
open access to it).

Downloads per dollar spent on OA vs TA just seems like an awkward way
of saying impact per dollar spent, especially since it's not the same
spender in the two cases!

> > BioMed Central: ~2000 downloads per year ( time of a paper); $500
> >processing charge per published paper
> > BioMed Central: cost per article use = $0.25, maximum
> Here BMC has calculated CPR (or rather cost-per-use) for the author by
> relating the fee paid by the author to the number of readings by readers.

When corrected to author-institution and user-institution, this can be
seen as download impact per dollar spent by institutions as a whole, in an
OA system vs. a TA system. This seems right, but again just a rather awkward
way of saying the OA is cheaper *and* generates more impact.

(I think the actual estimates of *how much cheaper*, based on BMC download
statistics, are probably somewhat arbitrary, because there are no good
download statistics for the comparison group. Hence citations are a more
reliable measure of impact. But I don't think anyone would dispute that
toll-free access generates more access, hence impact, than toll-based
access! Nor that $500 per OA article [BMC] paid by one author-institution
is less than the average $1500-$2000 per TA article paid jointly by all
subscribing user-institutions.)

> > Downloads are measured (by both OA-publishers and OAI archive-services),
> > not to track costs (which are irrelevant) but to measure research impact:
> Which is a good thing. My idea was to provide an additional measure
> cost-effectiveness (!) which would be comparable to the CPR-measure used by
> toll-access publishers.

But is it really an additional measure? Or just the impact measure translated
into funny-money?

> I did state in my article that: "The CPR measure relates the cost of
> resources to the amount of use and is therefore a measure of
> cost-effectiveness." and yet again "It might be worth restating that the CPR
> measure is a measure of cost-effectiveness and not a pay-per-view fee."


> > A good way to show the value of ads, but no relation to who's buying,
> > selling, or paying.
> This is what I meant! Now one can argue about the usefulness of such a
> measure...

I think researchers might find it more compelling to think of how many
downloads and citations their work can *earn* them, rather than how many
their institutional dollars can *buy* them...

Besides, OA can already be provided by researchers and their institutions
immediately, without having to change cost-recovery models or to estimate
the cost-effectiveness of such a change: They need only self-archive
their own TA articles. The enhanced "cost-effectiveness" comes then
even without any change in the cost! This "green road" to OA leaves it to
publishers to calculate the cost-effectiveness of eventually changing
to the OA cost-recovery model, if and when the availability of the OA
versions cause cancellation pressure on TA; meanwhile researchers and their
institutions are already long-since providing for one another's OA needs.

    "The Green Road to Open Access: A Leveraged Transition"

Stevan Harnad

NOTE: A complete archive of the ongoing discussion of providing open
access to the peer-reviewed research literature online (1998-2004)
is available at the American Scientist Open Access Forum:
        To join the Forum:
        Post discussion to:
        Hypermail Archive:

Unified Dual Open-Access-Provision Policy:
    BOAI-2 ("gold"): Publish your article in a suitable open-access
            journal whenever one exists.
    BOAI-1 ("green"): Otherwise, publish your article in a suitable
            toll-access journal and also self-archive it.

This message is sent to you because you are subscribed to
  The SPARC Open Access Forum.
To post, send your message to <>.
To unsubscribe, email to <>.
To switch to digest mode, email to <>.
To switch to index mode, email to <>.
Send administrative queries to <>.
Received on Tue Jan 20 2004 - 13:35:36 GMT

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:47:18 GMT