I have re-done the
UK calculation for Australia as follows (see this
article in
The Australian: Higher Education):
Australia is losing about $425 million dollars worth of potential return on its public investment in research every year. The Australian Research Councils spend about $1 billion dollars yearly, which generate about 32,000 research journal articles... The online-age practice of self-archiving has been shown to increase citation impact by a dramatic 50-250%, but so far only 15% of researchers are doing it...The marginal dollar value of one citation was estimated by Diamond in 1986 to range from $50-$1300 (US), depending on field and number of citations. (An increase from 0 to 1 citation is worth more than an increase from 30 to 31; most articles are in the citation range 0-5.) If we update by about 170% for inflation from 1986-2005 ($85.65-$2226.89) and convert from US dollars to Australian dollars, this yields the range $113-$2942 as the marginal value of an Australian citation to its author today. Self-archiving, as noted, increases citations by 50-250%, but, as also noted, only 15% of the articles being published are being self-archived today.
We will now apply only the most conservative ends of these estimates (50% citation increase from self-archiving at $113 per citation) to Australia’s current annual journal article output (and only for the approximately 32,000 Australian articles a year indexed by the Institute for Scientific Information, which covers only the top 8000 of the world's 24,000 journals). If we multiply by the 85% of Australia’s annual journal article output that is not yet self-archived (27,200 articles), this translates into an annual loss of $1, 536,800 in revenue to Australian researchers for not having done (or delegated) the few extra keystrokes per article it would have taken to self-archive their final drafts.
But this impact loss translates into a far bigger one for the Australian public, if we reckon it as the loss of potential returns on its research investment. As a proportion of Australia’s yearly $1 bn research expenditure (yielding 32,000 articles x 4.7 = 150,400 citations), our conservative estimate would be 50% x 85% x $1 bn = about $425 million dollars worth of loss in potential research impact (76,160 potential citations lost). And that is without even considering the wider loss in revenue from the loss of potential practical applications and usage of Australian research findings in Australia and worldwide, nor the still more general loss to the progress of human inquiry.
That has in turn elicited the following (anonymized) query:
"I'm having trouble reconciling the fact that the annual loss of $1,536,800 divided by the number of articles not archived (27,200) comes to $56.50 per article, half the stated value of a citation, $113... Does the following capture the meaning? (And if not, what do you mean?) "A 50 per cent citation increase added to a calculated $113 per citation, multiplied by the 85 per cent of articles not self-archived, multiplied by $1 billion in research funding gives a loss of potential returns of $425 million."
I'm afraid that's not it. But the actual connection between the two figures (the tax-payer's $425 million loss and the researchers' $1.5 million loss) is actually very simple, even if the explanation is a bit long-winded:
Very important: The estimate of the value of a citation to a researcher (based on the
Diamond data on citation revenue), totalling to a $1.5 million loss for the 85% of researchers who don't self-archive, is
not at all the same estimate as the $425 million loss for the Australian tax-payer. The bigger figure is based on
return on the tax-payer's investment, in terms of
citations. Citations are the measure of the tax-payer's value-for-money. Unused research is research that may as well not have been funded by the tax-payer: it's as if it had not been done at all. Citations are the measures of
usage: how much is the research I fund being applied and built upon?
So the best analogy on which to think of it is to think of the tax-payer as being like the buyer of a battery. A battery gets him 10 hours of usage. But evidence shows that if he refrigerates the battery before usage, he gets 50% more usage hours from the battery (15 hours). So if he pays $1 for the battery and doesn't refrigerate it, he only gets 10 hours of usage. He has lost 5 hours, or 50 cents' worth of potential usage for his money.
That's exactly the way not-self-archiving loses 50% of the potential usage (citations) on Australia's $1 billion research spend. (Actually $425 million, because about 15% of Australian researchers already do self-archive: 50% x 85% of the current citation total that $1 billion buys = $425 million's worth of citations.)
So if the Australian funding councils mandated self-archiving (as the
RCUK is planning to do in the UK), then that would increase the return on their research investment by $425 million.
Notice that the $113/56.50 revenue increase to researchers does not enter into this figure at all. That is a different matter. (If the ARC spend had been a less round figure than $1 billion, it would have been more obvious that the two have nothing to do with one another, apart from both being multiples of .5 and .85!)
If you want to see the connection between the return on the research investment to the research funder/tax-payer, which is reckoned in terms of the degree to which the research is actually put to use -- i.e., cited -- and the Diamond estimate of the worth of a single citation to a researcher, the way to understand it is this: The
reason we have the "publish or perish" mandate already is to make sure that the research we pay researchers to do is actually made available to be used and applied. So a much rougher estimate of the return on the research investment would be: number of publications. In the paper era, it was the number of publications that research employers (universities) and funders used to count and reward with salary increases etc. They were rewarding publishing in order to get what they really wanted, which was that the research should be used, applied, built-upon, so it could bring further benefits to the tax-payer funding it. Publish-or-perish was the incentive used to get the researchers to make their research known and available for use, and the number of articles published was the (crude) measure of the return on the research investment.
But today, in the online age, we see that publication-counting is too crude a measure of the return on the research investment (because work may be published, yet not used). What we need to reward is not just the number of publications, but the number of times they are
used (cited). Citation-counting had already started before the online era; that's why Diamond was already able, in 1986, to make his estimate of the amount that one citation is worth to a researcher. What has really changed since then with the increasing spread of the online medium is the possibility of
increasing the usage (citations) by 50%+ through self-archiving.
Hence to maximise the return (in terms of usage and citations) on the Australian investment in research, it is not enough to reward researchers for publishing (publish or perish), nor even to reward them per citation: self-archiving must be mandated. (It was possible to mandate "publish or perish." It is not possible to mandate "get-cited or perish," because the author cannot cite
himself [or rather, self-citations don't count]. But it
is possible to mandate "self-archive your publications or perish," because the author can self-archive them himself, thereby maximising citations!)
So in exchange for 100% self-archiving, Australia will get 50% (x 85%, since 15% already self-archive) more citations for the money it has spent on research ($425 million's worth more citations/usage) and Australian researchers will be rewarded for their pains (of self-archiving) by the revenue from 50% x $113 = $56.50 per article times 85% of the total of the 32,000 total article output (27,200) = $1.5 million more research revenue.
Does that make it clearer?
One more thing: The 50% increase in value-for-money for the Australian tax-payer (and to Australian research progress) -- $425 million's worth of citations -- as well as the 50% increase in reward to the Australian researcher ($56.50 per citation) are in part (not in all, but in part) a
competitive advantage, because the evidence on the 50%-250% increase in citations from self-archiving itself consists at least in part of a competitive advantage (a head start), based on comparing the number of citations for self-archived articles with the number of citations for non-self-archived articles in the same journal, same year.
Obviously, once self-archiving reaches 100%, the portion of the self-archiving advantage that is competitive -- the edge that the self-archived article has over its non-self-archived competitor -- will vanish. The playing field will be level. There will still be an overall advantage for the tax payer, because
all research will be accessible to
all potential users (and not just, as now, the research that each researcher's own institutions can afford to subscribe to), so it will all be used more, but only
some of it will be cited more (and some of it may even be cited less), for it will be merit and relevance alone that decide what research is used, unbiassed, as it is now, by affordability/accessibility.
It's been shown that even at 100% self-archiving, although researchers don't cite more than they did before,
downloads, for example, increase by 300%. Downloads are a measure of usage too: Before you can cite an article, you have to access it, and more downloads mean more accesses.
But overall, another incentive for Australia to mandate self-archiving now is that the race is to the swift and the battle to the strong (i.e., this is partly a horse-race): Those who mandate self-archiving
first gain a competitive advantage over those who do not. That increases the incentive, within Australia, for institutions to mandate self-archiving as soon as possible, for a bigger piece of the competitive pie, before it all levels out. And it also increases Australia's incentive to mandate self-archiving nationally, in its competition with other nations for research pre-eminence: If Australia were among the first to mandate self-archiving, it would give itself an immediate competitive boost, till other nations caught up.
But of course, research itself, worldwide, is the ultimate beneficiary of levelling the playing field, so the accessibility and hence the usage of research is maximised for the whole planet's research output.
(It is partly because of the competive ("head-start") advantage, which will shrink as we approach 100% self-archiving, that I used only the most conservative end of the observed 50%-250% citation advantage: 50%).
Stevan Harnad
